
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import pandas_datareader.data as web
import quandl
from datetime import datetime
import requests
import time
import warnings
warnings.filterwarnings('ignore')

from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Dense
from tensorflow.keras.optimizers import SGD,Adam
from keras.utils.np_utils import to_categorical
from tensorflow.keras.callbacks import EarlyStopping
from sklearn.preprocessing import MinMaxScaler

import tensorflow as tf
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Dense, Dropout, LSTM, BatchNormalization
from tensorflow.keras.callbacks import TensorBoard
from tensorflow.keras.callbacks import ModelCheckpoint



import pygad
import pygad.kerasga


FUTURE = 5
ROLLING = 5
ROLLING1 = 10
ROLLING2 = 15


#List of features

start = datetime(2005, 1,1)
end = datetime(2022,12,1)


inputs = [ 'NASDAQCOM', 
          'DEXJPUS', 'DEXUSEU', 'DEXCHUS', 'DEXUSAL', 'VIXCLS', 
          'DCOILWTICO']
fred_df = pd.DataFrame()

for i, name in enumerate(inputs):
    print(f'Series:{name}')
    if i == 0:
        fred_df = web.get_data_fred(name, start, end)
    else:
        _df = web.get_data_fred(name, start, end)
        fred_df = fred_df.join(_df, how = 'outer')

Series:NASDAQCOM
Series:DEXJPUS
Series:DEXUSEU
Series:DEXCHUS
Series:DEXUSAL
Series:VIXCLS
Series:DCOILWTICO


#Gold_NIFTY_Russell

gold = quandl.get("LBMA/GOLD", authtoken="kUWytaDzbhXbCX7PDzvg")
gold.index.name = 'DATE'
gold = gold[['USD (AM)']]
gold = gold.reset_index()

starttime = int(time.mktime(datetime(2005, 1,1).timetuple()))
endtime = int(time.mktime(datetime(2022, 12,1).timetuple()))
query_string = f'https://query1.finance.yahoo.com/v7/finance/download/^NSEI?period1={starttime}&period2={endtime}&interval=1d&events=history&includeAdjustedClose=true'
nifty = pd.read_csv(query_string)
nifty = nifty[['Date','Close' ]]
nifty.columns = ['DATE','NIFTY50' ]
nifty['DATE'] = pd.to_datetime(nifty['DATE'])

query_string = f'https://query1.finance.yahoo.com/v7/finance/download/^RUT?period1={starttime}&period2={endtime}&interval=1d&events=history&includeAdjustedClose=true'
russell = pd.read_csv(query_string)
russell = russell[['Date','Close' ]]
russell.columns = ['DATE','RUSSL' ]
russell['DATE'] = pd.to_datetime(russell['DATE'])

query_string = f'https://query1.finance.yahoo.com/v7/finance/download/^GSPC?period1={starttime}&period2={endtime}&interval=1d&events=history&includeAdjustedClose=true'
snp500 = pd.read_csv(query_string)
snp500 = snp500[['Date','Close' ]]
snp500.columns = ['DATE','SNP500' ]
snp500['DATE'] = pd.to_datetime(snp500['DATE'])


fred_df = fred_df.reset_index()


fred_df = pd.merge(pd.merge(fred_df, gold, on = 'DATE'), nifty, on = 'DATE')

fred_df = pd.merge(fred_df, snp500, on = 'DATE')


fred_df.isna().sum()

DATE           0
NASDAQCOM      0
DEXJPUS       39
DEXUSEU       39
DEXCHUS       39
DEXUSAL       39
VIXCLS         0
DCOILWTICO     7
USD (AM)       0
NIFTY50       10
SNP500         0
dtype: int64


fred_df['MONTH'] = fred_df['DATE'].dt.month
fred_df['DAY'] = fred_df['DATE'].dt.day


fred_df.columns = ['DATE', 'NASDAQCOM', 'DEXJPUS', 'DEXUSEU', 'DEXCHUS', 'DEXUSAL',
       'VIXCLS', 'DCOILWTICO', 'GOLD', 'NIFTY50', 'SNP500', 'MONTH',
       'DAY']

cols = ['NASDAQCOM', 'DEXJPUS', 'DEXUSEU', 'DEXCHUS', 'DEXUSAL',
       'VIXCLS', 'GOLD', 'NIFTY50', 'SNP500', 'DCOILWTICO']

fred_df.dropna(inplace = True)

for i in cols:
    col_name_roll = f'{i}_rolling_5'
    col_name_roll1 = f'{i}_rolling_10'
    col_name_roll2 = f'{i}_rolling_15'
    fred_df[col_name_roll] = fred_df[i].rolling(window = ROLLING).mean()
    fred_df[col_name_roll1] = fred_df[i].rolling(window = ROLLING1).mean()
    fred_df[col_name_roll2] = fred_df[i].rolling(window = ROLLING2).mean()
    
months = pd.get_dummies(fred_df['MONTH'], drop_first=True)
months.columns = ['FEB', 'MAR', 'APR', 'MAY', 'JUN', 'JUL', 'AUG', 'SEP', 'OCT', 'NOV', 'DEC']

day = pd.get_dummies(fred_df['DAY'], drop_first=True)

fred_df = pd.concat([fred_df, months, day], axis = 1)


#Before COVID
fred_df = fred_df[fred_df['DATE'] <= datetime(2020,3,1)]


fred_df['OIL_PRICE_FUTURE'] = fred_df['DCOILWTICO'].shift(-FUTURE)


del fred_df['DAY']
del fred_df['MONTH']


cols = ['NASDAQCOM',
                     'DEXJPUS',               'DEXUSEU',
                     'DEXCHUS',               'DEXUSAL',
                      'VIXCLS',            'DCOILWTICO',
                        'GOLD',               'NIFTY50',
                      'SNP500',   'NASDAQCOM_rolling_5',
        'NASDAQCOM_rolling_10',  'NASDAQCOM_rolling_15',
           'DEXJPUS_rolling_5',    'DEXJPUS_rolling_10',
          'DEXJPUS_rolling_15',     'DEXUSEU_rolling_5',
          'DEXUSEU_rolling_10',    'DEXUSEU_rolling_15',
           'DEXCHUS_rolling_5',    'DEXCHUS_rolling_10',
          'DEXCHUS_rolling_15',     'DEXUSAL_rolling_5',
          'DEXUSAL_rolling_10',    'DEXUSAL_rolling_15',
            'VIXCLS_rolling_5',     'VIXCLS_rolling_10',
           'VIXCLS_rolling_15',        'GOLD_rolling_5',
             'GOLD_rolling_10',       'GOLD_rolling_15',
           'NIFTY50_rolling_5',    'NIFTY50_rolling_10',
          'NIFTY50_rolling_15',      'SNP500_rolling_5',
           'SNP500_rolling_10',     'SNP500_rolling_15',
        'DCOILWTICO_rolling_5', 'DCOILWTICO_rolling_10',
       'DCOILWTICO_rolling_15']


minmaxscaler = MinMaxScaler()

fred_df[cols] = minmaxscaler.fit_transform(fred_df.loc[:, fred_df.columns.isin(cols)])

fred_df.dropna(inplace = True)


date_arr = fred_df['DATE']


fred_df.drop(columns='DATE', inplace = True)
fred_df = fred_df.reset_index()
del fred_df['index']


thres_hold = int(-0.10*(fred_df.shape[0]))

training_dataset = fred_df.iloc[:thres_hold, :]
test_dataset = fred_df.iloc[thres_hold:, :]


y_train = np.array(training_dataset['OIL_PRICE_FUTURE'].values)
y_test = np.array(test_dataset['OIL_PRICE_FUTURE'].values)

X_train = training_dataset.loc[:, training_dataset.columns != 'OIL_PRICE_FUTURE']
X_test = test_dataset.loc[:, test_dataset.columns != 'OIL_PRICE_FUTURE']

X_train_arr = np.array(X_train)
X_test_arr = np.array(X_test)


data_outputs = np.array(fred_df['OIL_PRICE_FUTURE'])
data_inputs = fred_df
del data_inputs['OIL_PRICE_FUTURE']


data_inputs = np.array(data_inputs)


#GA Implementation


num_samples = X_train.shape[0]
num_feature_elements = X_train.shape[1]
train_indices = np.arange(0, 2577, 1)
test_indices = np.arange(2577, 2863, 1)


sol_per_pop = 5
num_parents_mating = 4
num_mutations = 3

#Population Shape
pop_shape = (sol_per_pop, num_feature_elements)

#Initial Population
new_population = np.random.randint(low = 0, high = 2, size = pop_shape)
print(new_population.shape)

best_outputs = []
num_generations = 50

(5, 81)


pop_shape = (sol_per_pop, num_feature_elements)


def reduce_features(solution, features):
    
    selected_elements_indices = numpy.where(solution == 1)[0]
    reduced_features = features[:, selected_elements_indices]
    return reduced_features

def fitness_func(labels, predictions):
    mse = tf.keras.losses.MeanSquaredError()
    fitness = 1/(mse(labels, predictions).numpy() + 0.00001)
    return fitness

def cal_pop_fitness(pop, features, labels, train_indices, test_indices):
    accuracies = np.zeros(pop.shape[0])
    idx = 0
    for curr_solution in pop:
        reduced_features = reduce_features(curr_solution, features)
        train_data = reduced_features[train_indices, :]
        test_data = reduced_features[test_indices, :]
        train_labels = labels[train_indices]
        test_labels = labels[test_indices]
        print(test_data)
        print(train_data.shape)
        print(test_data.shape)
        model_nn = Sequential()
        model_nn.add(Dense(128, activation = 'sigmoid',input_shape = (train_data.shape[1],)))
        model_nn.add(Dropout(0.2))
        model_nn.add(BatchNormalization())
        model_nn.add(Dense(128, activation = 'relu'))
        model_nn.add(Dense(128, activation = 'relu'))
        model_nn.add(Dense(1, activation='relu'))
        opt = Adam()
        model_nn.compile(optimizer=opt,  loss='mean_squared_error', metrics='mean_squared_error')
        history_nn = model_nn.fit(train_data, train_labels, epochs = 25, validation_data = [test_data, test_labels])
        predictions = model_nn.predict(test_data)
        accuracies[idx] = fitness_func(test_labels, predictions)
        idx = idx + 1
    return accuracies

def select_mating_pool(pop, fitness, num_parents):
    parents = numpy.empty((num_parents, pop_shape[1]))
    for parent_num in range(num_parents):
        max_fitness_idx = np.where(fitness == np.max(fitness))
        max_fitness_idx = max_fitness_idx[0][0]
        parents[parent_num, :] = pop[max_fitness_idx, :]
        fitness[max_fitness_idx] = -99999999999
    return parents

def crossover(parents, offspring_size):
    offspring = numpy.empty(offspring_size)
    # The point at which crossover takes place between two parents. Usually, it is at the center.
    crossover_point = numpy.uint8(offspring_size[1]/2)

    for k in range(offspring_size[0]):
        # Index of the first parent to mate.
        parent1_idx = k%parents.shape[0]
        # Index of the second parent to mate.
        parent2_idx = (k+1)%parents.shape[0]
        # The new offspring will have its first half of its genes taken from the first parent.
        offspring[k, 0:crossover_point] = parents[parent1_idx, 0:crossover_point]
        # The new offspring will have its second half of its genes taken from the second parent.
        offspring[k, crossover_point:] = parents[parent2_idx, crossover_point:]
    return offspring


def mutation(offspring_crossover, num_mutations = 2):
    mutation_idx = np.random.randint(low = 0, high = offspring_crossover.shape[1], size = num_mutations)
    for idx in range(offspring_crossover.shape[0]):
        offspring_crossover[idx, mutation_idx] = 1 - offspring_crossover[idx, mutation_idx]
    return offspring_crossover


for generation in range(num_generations):
    print("Generation: ", generation)
    fitness = cal_pop_fitness(new_population, data_inputs, data_outputs, train_indices, test_indices)
    best_outputs.append(np.max(fitness))
    print("Best Result: ", best_outputs[-1])

    parents = select_mating_pool(new_population, fitness, num_parents_mating)
    offspring_crossover = crossover(parents, offspring_size= (pop_shape[0]-parents.shape[0], num_feature_elements))
    offspring_mutation = mutation(offspring_crossover, num_mutations= num_mutations)
    
    new_population[0:parents.shape[0], :] = parents
    new_population[parents.shape[0]:, :] = offspring_mutation


fitness = cal_pop_fitness(new_population, data_inputs, data_outputs, train_indices, test_indices)
# Then return the index of that solution corresponding to the best fitness.
best_match_idx = numpy.where(fitness == numpy.max(fitness))[0]
best_match_idx = best_match_idx[0]

best_solution = new_population[best_match_idx, :]
best_solution_indices = numpy.where(best_solution == 1)[0]
best_solution_num_elements = best_solution_indices.shape[0]
best_solution_fitness = fitness[best_match_idx]

print("best_match_idx : ", best_match_idx)
print("best_solution : ", best_solution)
print("Selected indices : ", best_solution_indices)
print("Number of selected elements : ", best_solution_num_elements)
print("Best solution fitness : ", best_solution_fitness)

Generation:  0
[[0.6994447  0.18225377 0.25015142 ... 0.         0.         0.        ]
 [0.6737802  0.19041704 0.24591157 ... 0.         0.         0.        ]
 [0.65978331 0.18030169 0.23682617 ... 0.         0.         0.        ]
 ...
 [0.99012252 0.07808341 0.12517666 ... 0.         0.         0.        ]
 [1.         0.0743567  0.1211387  ... 0.         0.         0.        ]
 [0.99225423 0.07488909 0.11003432 ... 0.         0.         0.        ]]
(2577, 41)
(286, 41)
Epoch 1/25

2022-12-09 11:08:06.783855: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.

81/81 [==============================] - ETA: 0s - loss: 2390.3926 - mean_squared_error: 2390.3926

2022-12-09 11:08:12.007689: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.

81/81 [==============================] - 6s 58ms/step - loss: 2390.3926 - mean_squared_error: 2390.3926 - val_loss: 818.6481 - val_mean_squared_error: 818.6481
Epoch 2/25
81/81 [==============================] - 1s 14ms/step - loss: 121.0158 - mean_squared_error: 121.0158 - val_loss: 1439.9713 - val_mean_squared_error: 1439.9713
Epoch 3/25
81/81 [==============================] - 1s 14ms/step - loss: 90.5931 - mean_squared_error: 90.5931 - val_loss: 2317.9133 - val_mean_squared_error: 2317.9133
Epoch 4/25
81/81 [==============================] - 1s 10ms/step - loss: 81.2363 - mean_squared_error: 81.2363 - val_loss: 2625.4529 - val_mean_squared_error: 2625.4529
Epoch 5/25
81/81 [==============================] - 1s 10ms/step - loss: 68.6132 - mean_squared_error: 68.6132 - val_loss: 2683.2451 - val_mean_squared_error: 2683.2451
Epoch 6/25
81/81 [==============================] - 1s 10ms/step - loss: 56.2447 - mean_squared_error: 56.2447 - val_loss: 2638.3518 - val_mean_squared_error: 2638.3518
Epoch 7/25
81/81 [==============================] - 1s 8ms/step - loss: 46.3267 - mean_squared_error: 46.3267 - val_loss: 2724.8137 - val_mean_squared_error: 2724.8137
Epoch 8/25
81/81 [==============================] - 1s 9ms/step - loss: 47.0067 - mean_squared_error: 47.0067 - val_loss: 2641.4404 - val_mean_squared_error: 2641.4404
Epoch 9/25
81/81 [==============================] - 1s 8ms/step - loss: 44.2807 - mean_squared_error: 44.2807 - val_loss: 2646.6050 - val_mean_squared_error: 2646.6050
Epoch 10/25
81/81 [==============================] - 1s 10ms/step - loss: 45.7578 - mean_squared_error: 45.7578 - val_loss: 2620.6013 - val_mean_squared_error: 2620.6013
Epoch 11/25
81/81 [==============================] - 1s 9ms/step - loss: 45.9175 - mean_squared_error: 45.9175 - val_loss: 2574.8103 - val_mean_squared_error: 2574.8103
Epoch 12/25
81/81 [==============================] - 1s 9ms/step - loss: 52.7963 - mean_squared_error: 52.7963 - val_loss: 2468.1794 - val_mean_squared_error: 2468.1794
Epoch 13/25
81/81 [==============================] - 1s 9ms/step - loss: 42.9988 - mean_squared_error: 42.9988 - val_loss: 2551.1704 - val_mean_squared_error: 2551.1704
Epoch 14/25
81/81 [==============================] - 1s 8ms/step - loss: 48.5752 - mean_squared_error: 48.5752 - val_loss: 2464.7446 - val_mean_squared_error: 2464.7446
Epoch 15/25
81/81 [==============================] - 1s 8ms/step - loss: 43.5355 - mean_squared_error: 43.5355 - val_loss: 2427.3105 - val_mean_squared_error: 2427.3103
Epoch 16/25
81/81 [==============================] - 1s 8ms/step - loss: 48.5332 - mean_squared_error: 48.5332 - val_loss: 2375.1843 - val_mean_squared_error: 2375.1843
Epoch 17/25
81/81 [==============================] - 1s 8ms/step - loss: 45.3841 - mean_squared_error: 45.3841 - val_loss: 2278.3857 - val_mean_squared_error: 2278.3855
Epoch 18/25
81/81 [==============================] - 1s 9ms/step - loss: 44.7125 - mean_squared_error: 44.7125 - val_loss: 2333.3694 - val_mean_squared_error: 2333.3694
Epoch 19/25
81/81 [==============================] - 1s 8ms/step - loss: 45.7382 - mean_squared_error: 45.7382 - val_loss: 2324.3188 - val_mean_squared_error: 2324.3188
Epoch 20/25
81/81 [==============================] - 1s 8ms/step - loss: 38.3040 - mean_squared_error: 38.3040 - val_loss: 2199.1931 - val_mean_squared_error: 2199.1931
Epoch 21/25
81/81 [==============================] - 1s 8ms/step - loss: 41.8382 - mean_squared_error: 41.8382 - val_loss: 2277.7341 - val_mean_squared_error: 2277.7339
Epoch 22/25
81/81 [==============================] - 1s 8ms/step - loss: 41.6554 - mean_squared_error: 41.6554 - val_loss: 2145.1177 - val_mean_squared_error: 2145.1177
Epoch 23/25
81/81 [==============================] - 1s 8ms/step - loss: 42.0652 - mean_squared_error: 42.0652 - val_loss: 2269.1819 - val_mean_squared_error: 2269.1819
Epoch 24/25
81/81 [==============================] - 1s 8ms/step - loss: 48.1477 - mean_squared_error: 48.1477 - val_loss: 2276.0862 - val_mean_squared_error: 2276.0862
Epoch 25/25
81/81 [==============================] - 1s 8ms/step - loss: 42.4544 - mean_squared_error: 42.4544 - val_loss: 2167.9163 - val_mean_squared_error: 2167.9163

2022-12-09 11:08:30.723773: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.

[[0.6994447  0.82922519 0.76015828 ... 0.         0.         0.        ]
 [0.6737802  0.83747884 0.74335148 ... 0.         0.         0.        ]
 [0.65978331 0.82658241 0.72532678 ... 0.         0.         0.        ]
 ...
 [0.99012252 0.96241146 0.99414683 ... 0.         0.         0.        ]
 [1.         0.97597103 1.         ... 0.         0.         0.        ]
 [0.99225423 0.97139181 0.99523183 ... 0.         0.         0.        ]]
(2577, 41)
(286, 41)
Epoch 1/25

2022-12-09 11:08:31.365523: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.

81/81 [==============================] - ETA: 0s - loss: 2386.3457 - mean_squared_error: 2386.3457

2022-12-09 11:08:36.346311: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.


reduced_data_inputs = data_inputs[:, [   0 , 5 , 7,  9, 10, 11, 14 ,17, 20, 21, 26, 28, 30, 33, 34, 35, 36, 47, 48, 49, 50, 51, 53, 54,
 58, 60, 62, 63, 65, 66, 67, 68, 69, 70, 74, 75, 76, 77, 78, 79, 80]]


train_data_ann = reduced_data_inputs[train_indices, :]
test_data_ann = reduced_data_inputs[test_indices, :]


train_data_ann.shape

(2577, 41)


y_train_ann = data_outputs[train_indices]

y_test_ann = data_outputs[test_indices]


model_nn = Sequential()
model_nn.add(Dense(128, activation = 'sigmoid',input_shape = (train_data_ann.shape[1],)))
model_nn.add(Dropout(0.2))
model_nn.add(BatchNormalization())
model_nn.add(Dense(128, activation = 'relu'))
model_nn.add(Dense(128, activation = 'relu'))
model_nn.add(Dense(1, activation='relu'))

opt = Adam()

model_nn.compile(optimizer=opt,  loss='mean_squared_error', metrics='mean_squared_error')
history_nn = model_nn.fit(train_data_ann, y_train_ann, epochs = 100, validation_data = [test_data_ann, y_test_ann])

2022-12-09 16:43:09.995501: W tensorflow/core/platform/profile_utils/cpu_utils.cc:128] Failed to get CPU frequency: 0 Hz
2022-12-09 16:43:10.145530: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.

Epoch 1/100
81/81 [==============================] - 1s 9ms/step - loss: 6334.6709 - mean_squared_error: 6334.6709 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 2/100
 8/81 [=>............................] - ETA: 0s - loss: 6440.0654 - mean_squared_error: 6440.0654

2022-12-09 16:43:11.052232: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.

81/81 [==============================] - 1s 8ms/step - loss: 6334.6724 - mean_squared_error: 6334.6724 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 3/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6719 - mean_squared_error: 6334.6719 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 4/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6729 - mean_squared_error: 6334.6729 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 5/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6719 - mean_squared_error: 6334.6719 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 6/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6714 - mean_squared_error: 6334.6714 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 7/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6704 - mean_squared_error: 6334.6704 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 8/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6729 - mean_squared_error: 6334.6729 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 9/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6714 - mean_squared_error: 6334.6714 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 10/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6719 - mean_squared_error: 6334.6719 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 11/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6709 - mean_squared_error: 6334.6709 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 12/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6724 - mean_squared_error: 6334.6724 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 13/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6719 - mean_squared_error: 6334.6719 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 14/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6719 - mean_squared_error: 6334.6719 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 15/100
81/81 [==============================] - 1s 7ms/step - loss: 6334.6714 - mean_squared_error: 6334.6714 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 16/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6724 - mean_squared_error: 6334.6724 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 17/100
81/81 [==============================] - 1s 7ms/step - loss: 6334.6714 - mean_squared_error: 6334.6714 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 18/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6704 - mean_squared_error: 6334.6704 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 19/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6724 - mean_squared_error: 6334.6724 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 20/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6719 - mean_squared_error: 6334.6719 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 21/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6719 - mean_squared_error: 6334.6719 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 22/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6704 - mean_squared_error: 6334.6704 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 23/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6704 - mean_squared_error: 6334.6704 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 24/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6724 - mean_squared_error: 6334.6724 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 25/100
81/81 [==============================] - 1s 7ms/step - loss: 6334.6724 - mean_squared_error: 6334.6724 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 26/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6719 - mean_squared_error: 6334.6719 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 27/100
81/81 [==============================] - 1s 7ms/step - loss: 6334.6724 - mean_squared_error: 6334.6724 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 28/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6714 - mean_squared_error: 6334.6714 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 29/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6733 - mean_squared_error: 6334.6733 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 30/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6724 - mean_squared_error: 6334.6724 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 31/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6704 - mean_squared_error: 6334.6704 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 32/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6733 - mean_squared_error: 6334.6733 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 33/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6724 - mean_squared_error: 6334.6724 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 34/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6709 - mean_squared_error: 6334.6709 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 35/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6719 - mean_squared_error: 6334.6719 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 36/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6724 - mean_squared_error: 6334.6724 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 37/100
81/81 [==============================] - 1s 7ms/step - loss: 6334.6724 - mean_squared_error: 6334.6724 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 38/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6714 - mean_squared_error: 6334.6714 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 39/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6709 - mean_squared_error: 6334.6709 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 40/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6709 - mean_squared_error: 6334.6709 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 41/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6724 - mean_squared_error: 6334.6724 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 42/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6719 - mean_squared_error: 6334.6719 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 43/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6724 - mean_squared_error: 6334.6724 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 44/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6709 - mean_squared_error: 6334.6709 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 45/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6714 - mean_squared_error: 6334.6714 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 46/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6729 - mean_squared_error: 6334.6729 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 47/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6709 - mean_squared_error: 6334.6709 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 48/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6724 - mean_squared_error: 6334.6724 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 49/100
81/81 [==============================] - 1s 7ms/step - loss: 6334.6704 - mean_squared_error: 6334.6704 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 50/100
81/81 [==============================] - 1s 7ms/step - loss: 6334.6704 - mean_squared_error: 6334.6704 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 51/100
81/81 [==============================] - 1s 7ms/step - loss: 6334.6724 - mean_squared_error: 6334.6724 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 52/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6719 - mean_squared_error: 6334.6719 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 53/100
81/81 [==============================] - 1s 7ms/step - loss: 6334.6724 - mean_squared_error: 6334.6724 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 54/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6709 - mean_squared_error: 6334.6709 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 55/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6709 - mean_squared_error: 6334.6709 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 56/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6724 - mean_squared_error: 6334.6724 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 57/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6729 - mean_squared_error: 6334.6729 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 58/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6724 - mean_squared_error: 6334.6724 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 59/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6709 - mean_squared_error: 6334.6709 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 60/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6714 - mean_squared_error: 6334.6714 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 61/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6724 - mean_squared_error: 6334.6724 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 62/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6714 - mean_squared_error: 6334.6714 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 63/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6724 - mean_squared_error: 6334.6724 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 64/100
81/81 [==============================] - 1s 8ms/step - loss: 6334.6724 - mean_squared_error: 6334.6724 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 65/100
81/81 [==============================] - 1s 8ms/step - loss: 6322.5117 - mean_squared_error: 6322.5117 - val_loss: 2989.6125 - val_mean_squared_error: 2989.6125
Epoch 66/100
81/81 [==============================] - 1s 8ms/step - loss: 702.4682 - mean_squared_error: 702.4682 - val_loss: 2562.2031 - val_mean_squared_error: 2562.2031
Epoch 67/100
81/81 [==============================] - 1s 7ms/step - loss: 167.7946 - mean_squared_error: 167.7946 - val_loss: 2079.3723 - val_mean_squared_error: 2079.3723
Epoch 68/100
81/81 [==============================] - 1s 7ms/step - loss: 147.1697 - mean_squared_error: 147.1697 - val_loss: 2258.4800 - val_mean_squared_error: 2258.4800
Epoch 69/100
81/81 [==============================] - 1s 8ms/step - loss: 142.5865 - mean_squared_error: 142.5865 - val_loss: 2051.7507 - val_mean_squared_error: 2051.7505
Epoch 70/100
81/81 [==============================] - 1s 8ms/step - loss: 138.2778 - mean_squared_error: 138.2778 - val_loss: 2124.7263 - val_mean_squared_error: 2124.7263
Epoch 71/100
81/81 [==============================] - 1s 8ms/step - loss: 129.6876 - mean_squared_error: 129.6876 - val_loss: 2064.8970 - val_mean_squared_error: 2064.8970
Epoch 72/100
81/81 [==============================] - 1s 8ms/step - loss: 129.3446 - mean_squared_error: 129.3446 - val_loss: 1944.3743 - val_mean_squared_error: 1944.3743
Epoch 73/100
81/81 [==============================] - 1s 8ms/step - loss: 132.5877 - mean_squared_error: 132.5877 - val_loss: 1835.8247 - val_mean_squared_error: 1835.8247
Epoch 74/100
81/81 [==============================] - 1s 8ms/step - loss: 123.4942 - mean_squared_error: 123.4942 - val_loss: 1746.5275 - val_mean_squared_error: 1746.5275
Epoch 75/100
81/81 [==============================] - 1s 8ms/step - loss: 118.3493 - mean_squared_error: 118.3493 - val_loss: 1602.7944 - val_mean_squared_error: 1602.7944
Epoch 76/100
81/81 [==============================] - 1s 8ms/step - loss: 121.2151 - mean_squared_error: 121.2151 - val_loss: 1628.8861 - val_mean_squared_error: 1628.8861
Epoch 77/100
81/81 [==============================] - 1s 8ms/step - loss: 115.1444 - mean_squared_error: 115.1444 - val_loss: 1258.4623 - val_mean_squared_error: 1258.4623
Epoch 78/100
81/81 [==============================] - 1s 7ms/step - loss: 115.0989 - mean_squared_error: 115.0989 - val_loss: 1195.8456 - val_mean_squared_error: 1195.8456
Epoch 79/100
81/81 [==============================] - 1s 8ms/step - loss: 109.2188 - mean_squared_error: 109.2188 - val_loss: 1145.2429 - val_mean_squared_error: 1145.2429
Epoch 80/100
81/81 [==============================] - 1s 8ms/step - loss: 115.3049 - mean_squared_error: 115.3049 - val_loss: 1053.3711 - val_mean_squared_error: 1053.3711
Epoch 81/100
81/81 [==============================] - 1s 8ms/step - loss: 112.9429 - mean_squared_error: 112.9429 - val_loss: 1149.2200 - val_mean_squared_error: 1149.2200
Epoch 82/100
81/81 [==============================] - 1s 8ms/step - loss: 100.8549 - mean_squared_error: 100.8549 - val_loss: 992.7377 - val_mean_squared_error: 992.7377
Epoch 83/100
81/81 [==============================] - 1s 8ms/step - loss: 102.4521 - mean_squared_error: 102.4521 - val_loss: 883.2311 - val_mean_squared_error: 883.2311
Epoch 84/100
81/81 [==============================] - 1s 8ms/step - loss: 103.7124 - mean_squared_error: 103.7124 - val_loss: 1074.7314 - val_mean_squared_error: 1074.7314
Epoch 85/100
81/81 [==============================] - 1s 8ms/step - loss: 96.5222 - mean_squared_error: 96.5222 - val_loss: 871.7490 - val_mean_squared_error: 871.7490
Epoch 86/100
81/81 [==============================] - 1s 8ms/step - loss: 99.2997 - mean_squared_error: 99.2997 - val_loss: 788.4041 - val_mean_squared_error: 788.4041
Epoch 87/100
81/81 [==============================] - 1s 8ms/step - loss: 100.9195 - mean_squared_error: 100.9195 - val_loss: 709.0814 - val_mean_squared_error: 709.0814
Epoch 88/100
81/81 [==============================] - 1s 8ms/step - loss: 98.0730 - mean_squared_error: 98.0730 - val_loss: 612.5357 - val_mean_squared_error: 612.5357
Epoch 89/100
81/81 [==============================] - 1s 8ms/step - loss: 89.6764 - mean_squared_error: 89.6764 - val_loss: 627.9725 - val_mean_squared_error: 627.9725
Epoch 90/100
81/81 [==============================] - 1s 8ms/step - loss: 87.5772 - mean_squared_error: 87.5772 - val_loss: 853.0435 - val_mean_squared_error: 853.0435
Epoch 91/100
81/81 [==============================] - 1s 8ms/step - loss: 87.7942 - mean_squared_error: 87.7942 - val_loss: 704.4908 - val_mean_squared_error: 704.4908
Epoch 92/100
81/81 [==============================] - 1s 8ms/step - loss: 84.9635 - mean_squared_error: 84.9635 - val_loss: 674.9868 - val_mean_squared_error: 674.9868
Epoch 93/100
81/81 [==============================] - 1s 8ms/step - loss: 84.6088 - mean_squared_error: 84.6088 - val_loss: 601.9120 - val_mean_squared_error: 601.9120
Epoch 94/100
81/81 [==============================] - 1s 8ms/step - loss: 79.7513 - mean_squared_error: 79.7513 - val_loss: 742.5476 - val_mean_squared_error: 742.5476
Epoch 95/100
81/81 [==============================] - 1s 8ms/step - loss: 79.8271 - mean_squared_error: 79.8271 - val_loss: 590.5106 - val_mean_squared_error: 590.5106
Epoch 96/100
81/81 [==============================] - 1s 8ms/step - loss: 84.8335 - mean_squared_error: 84.8335 - val_loss: 553.4356 - val_mean_squared_error: 553.4356
Epoch 97/100
81/81 [==============================] - 1s 7ms/step - loss: 77.4249 - mean_squared_error: 77.4249 - val_loss: 459.4814 - val_mean_squared_error: 459.4814
Epoch 98/100
81/81 [==============================] - 1s 7ms/step - loss: 80.3979 - mean_squared_error: 80.3979 - val_loss: 515.1418 - val_mean_squared_error: 515.1418
Epoch 99/100
81/81 [==============================] - 1s 8ms/step - loss: 78.1820 - mean_squared_error: 78.1820 - val_loss: 397.8671 - val_mean_squared_error: 397.8671
Epoch 100/100
81/81 [==============================] - 1s 8ms/step - loss: 76.9606 - mean_squared_error: 76.9606 - val_loss: 344.2016 - val_mean_squared_error: 344.2016


y_pred__test_ga_nn = model_nn.predict(test_data_ann)
y_pred__train_ga_nn = model_nn.predict(train_data_ann)

2022-12-09 16:44:16.813375: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.


#Main Architechture ANN
model_nn = Sequential()
model_nn.add(Dense(128, activation = 'sigmoid',input_shape = (X_train.shape[1],)))
model_nn.add(Dropout(0.2))
model_nn.add(BatchNormalization())
model_nn.add(Dense(128, activation = 'relu'))
model_nn.add(Dense(128, activation = 'relu'))
model_nn.add(Dense(1, activation='relu'))

opt = Adam()

model_nn.compile(optimizer=opt,  loss='mean_squared_error', metrics='mean_squared_error')

history_nn = model_nn.fit(X_train, y_train, epochs = 100, 
             validation_data = [X_test, y_test])

Epoch 1/100
 1/81 [..............................] - ETA: 25s - loss: 5526.8862 - mean_squared_error: 5526.8862

2022-12-09 17:17:10.563268: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.

81/81 [==============================] - 1s 10ms/step - loss: 6334.6724 - mean_squared_error: 6334.6724 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 2/100
 8/81 [=>............................] - ETA: 0s - loss: 6339.9668 - mean_squared_error: 6339.9668

2022-12-09 17:17:11.442433: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.

81/81 [==============================] - 1s 8ms/step - loss: 6334.6631 - mean_squared_error: 6334.6631 - val_loss: 3164.6384 - val_mean_squared_error: 3164.6384
Epoch 3/100
81/81 [==============================] - 1s 8ms/step - loss: 1747.6044 - mean_squared_error: 1747.6044 - val_loss: 1564.3827 - val_mean_squared_error: 1564.3827
Epoch 4/100
81/81 [==============================] - 1s 8ms/step - loss: 72.3302 - mean_squared_error: 72.3302 - val_loss: 1900.6294 - val_mean_squared_error: 1900.6294
Epoch 5/100
81/81 [==============================] - 1s 8ms/step - loss: 54.8067 - mean_squared_error: 54.8067 - val_loss: 2106.8301 - val_mean_squared_error: 2106.8301
Epoch 6/100
81/81 [==============================] - 1s 8ms/step - loss: 49.1724 - mean_squared_error: 49.1724 - val_loss: 2147.6216 - val_mean_squared_error: 2147.6216
Epoch 7/100
81/81 [==============================] - 1s 8ms/step - loss: 42.7373 - mean_squared_error: 42.7373 - val_loss: 2233.4321 - val_mean_squared_error: 2233.4321
Epoch 8/100
81/81 [==============================] - 1s 8ms/step - loss: 39.3994 - mean_squared_error: 39.3994 - val_loss: 2281.6738 - val_mean_squared_error: 2281.6738
Epoch 9/100
81/81 [==============================] - 1s 8ms/step - loss: 38.6420 - mean_squared_error: 38.6420 - val_loss: 2193.5852 - val_mean_squared_error: 2193.5852
Epoch 10/100
81/81 [==============================] - 1s 8ms/step - loss: 38.8651 - mean_squared_error: 38.8651 - val_loss: 2209.5571 - val_mean_squared_error: 2209.5571
Epoch 11/100
81/81 [==============================] - 1s 8ms/step - loss: 43.0868 - mean_squared_error: 43.0868 - val_loss: 2266.3347 - val_mean_squared_error: 2266.3347
Epoch 12/100
81/81 [==============================] - 1s 8ms/step - loss: 40.9041 - mean_squared_error: 40.9041 - val_loss: 2144.6250 - val_mean_squared_error: 2144.6250
Epoch 13/100
81/81 [==============================] - 1s 8ms/step - loss: 38.9563 - mean_squared_error: 38.9563 - val_loss: 2029.2386 - val_mean_squared_error: 2029.2386
Epoch 14/100
81/81 [==============================] - 1s 8ms/step - loss: 30.6986 - mean_squared_error: 30.6986 - val_loss: 2095.4822 - val_mean_squared_error: 2095.4822
Epoch 15/100
81/81 [==============================] - 1s 8ms/step - loss: 39.9870 - mean_squared_error: 39.9870 - val_loss: 2070.7480 - val_mean_squared_error: 2070.7480
Epoch 16/100
81/81 [==============================] - 1s 8ms/step - loss: 41.4760 - mean_squared_error: 41.4760 - val_loss: 1935.4486 - val_mean_squared_error: 1935.4486
Epoch 17/100
81/81 [==============================] - 1s 8ms/step - loss: 40.0728 - mean_squared_error: 40.0728 - val_loss: 2160.9487 - val_mean_squared_error: 2160.9487
Epoch 18/100
81/81 [==============================] - 1s 8ms/step - loss: 34.1476 - mean_squared_error: 34.1476 - val_loss: 1975.4473 - val_mean_squared_error: 1975.4473
Epoch 19/100
81/81 [==============================] - 1s 8ms/step - loss: 44.3044 - mean_squared_error: 44.3044 - val_loss: 1869.0446 - val_mean_squared_error: 1869.0446
Epoch 20/100
81/81 [==============================] - 1s 8ms/step - loss: 34.2134 - mean_squared_error: 34.2134 - val_loss: 1812.9460 - val_mean_squared_error: 1812.9460
Epoch 21/100
81/81 [==============================] - 1s 8ms/step - loss: 35.7597 - mean_squared_error: 35.7597 - val_loss: 1726.0461 - val_mean_squared_error: 1726.0461
Epoch 22/100
81/81 [==============================] - 1s 8ms/step - loss: 36.2443 - mean_squared_error: 36.2443 - val_loss: 1707.1045 - val_mean_squared_error: 1707.1045
Epoch 23/100
81/81 [==============================] - 1s 8ms/step - loss: 32.3756 - mean_squared_error: 32.3756 - val_loss: 1682.9875 - val_mean_squared_error: 1682.9875
Epoch 24/100
81/81 [==============================] - 1s 8ms/step - loss: 34.5654 - mean_squared_error: 34.5654 - val_loss: 1687.9838 - val_mean_squared_error: 1687.9838
Epoch 25/100
81/81 [==============================] - 1s 8ms/step - loss: 35.6989 - mean_squared_error: 35.6989 - val_loss: 1728.9198 - val_mean_squared_error: 1728.9198
Epoch 26/100
81/81 [==============================] - 1s 8ms/step - loss: 35.3952 - mean_squared_error: 35.3952 - val_loss: 1660.1007 - val_mean_squared_error: 1660.1007
Epoch 27/100
81/81 [==============================] - 1s 8ms/step - loss: 36.5092 - mean_squared_error: 36.5092 - val_loss: 1418.7063 - val_mean_squared_error: 1418.7063
Epoch 28/100
81/81 [==============================] - 1s 8ms/step - loss: 37.8989 - mean_squared_error: 37.8989 - val_loss: 1758.6995 - val_mean_squared_error: 1758.6995
Epoch 29/100
81/81 [==============================] - 1s 8ms/step - loss: 42.9441 - mean_squared_error: 42.9441 - val_loss: 1437.5389 - val_mean_squared_error: 1437.5389
Epoch 30/100
81/81 [==============================] - 1s 8ms/step - loss: 36.9662 - mean_squared_error: 36.9662 - val_loss: 1391.8309 - val_mean_squared_error: 1391.8309
Epoch 31/100
81/81 [==============================] - 1s 8ms/step - loss: 29.7418 - mean_squared_error: 29.7418 - val_loss: 1576.3597 - val_mean_squared_error: 1576.3597
Epoch 32/100
81/81 [==============================] - 1s 8ms/step - loss: 30.6758 - mean_squared_error: 30.6758 - val_loss: 1565.0532 - val_mean_squared_error: 1565.0532
Epoch 33/100
81/81 [==============================] - 1s 8ms/step - loss: 35.8518 - mean_squared_error: 35.8518 - val_loss: 1538.6613 - val_mean_squared_error: 1538.6613
Epoch 34/100
81/81 [==============================] - 1s 8ms/step - loss: 36.5330 - mean_squared_error: 36.5330 - val_loss: 1526.6255 - val_mean_squared_error: 1526.6255
Epoch 35/100
81/81 [==============================] - 1s 8ms/step - loss: 32.4391 - mean_squared_error: 32.4391 - val_loss: 1496.3749 - val_mean_squared_error: 1496.3749
Epoch 36/100
81/81 [==============================] - 1s 8ms/step - loss: 33.8167 - mean_squared_error: 33.8167 - val_loss: 1518.7776 - val_mean_squared_error: 1518.7776
Epoch 37/100
81/81 [==============================] - 1s 8ms/step - loss: 30.2128 - mean_squared_error: 30.2128 - val_loss: 1355.7388 - val_mean_squared_error: 1355.7386
Epoch 38/100
81/81 [==============================] - 1s 8ms/step - loss: 33.5150 - mean_squared_error: 33.5150 - val_loss: 1487.9640 - val_mean_squared_error: 1487.9640
Epoch 39/100
81/81 [==============================] - 1s 8ms/step - loss: 34.2013 - mean_squared_error: 34.2013 - val_loss: 1399.2528 - val_mean_squared_error: 1399.2528
Epoch 40/100
81/81 [==============================] - 1s 8ms/step - loss: 28.7668 - mean_squared_error: 28.7668 - val_loss: 1366.4402 - val_mean_squared_error: 1366.4402
Epoch 41/100
81/81 [==============================] - 1s 8ms/step - loss: 34.1814 - mean_squared_error: 34.1814 - val_loss: 1329.1292 - val_mean_squared_error: 1329.1292
Epoch 42/100
81/81 [==============================] - 1s 8ms/step - loss: 28.8237 - mean_squared_error: 28.8237 - val_loss: 1263.3940 - val_mean_squared_error: 1263.3940
Epoch 43/100
81/81 [==============================] - 1s 8ms/step - loss: 35.4947 - mean_squared_error: 35.4947 - val_loss: 1455.4680 - val_mean_squared_error: 1455.4680
Epoch 44/100
81/81 [==============================] - 1s 8ms/step - loss: 26.6467 - mean_squared_error: 26.6467 - val_loss: 1185.8441 - val_mean_squared_error: 1185.8441
Epoch 45/100
81/81 [==============================] - 1s 8ms/step - loss: 34.6517 - mean_squared_error: 34.6517 - val_loss: 1350.9700 - val_mean_squared_error: 1350.9700
Epoch 46/100
81/81 [==============================] - 1s 8ms/step - loss: 33.0527 - mean_squared_error: 33.0527 - val_loss: 1237.7695 - val_mean_squared_error: 1237.7695
Epoch 47/100
81/81 [==============================] - 1s 8ms/step - loss: 29.8159 - mean_squared_error: 29.8159 - val_loss: 1159.9984 - val_mean_squared_error: 1159.9984
Epoch 48/100
81/81 [==============================] - 1s 8ms/step - loss: 35.0053 - mean_squared_error: 35.0053 - val_loss: 1678.6833 - val_mean_squared_error: 1678.6833
Epoch 49/100
81/81 [==============================] - 1s 9ms/step - loss: 27.9078 - mean_squared_error: 27.9078 - val_loss: 1256.7440 - val_mean_squared_error: 1256.7440
Epoch 50/100
81/81 [==============================] - 1s 8ms/step - loss: 27.1958 - mean_squared_error: 27.1958 - val_loss: 1234.7739 - val_mean_squared_error: 1234.7739
Epoch 51/100
81/81 [==============================] - 1s 8ms/step - loss: 30.3254 - mean_squared_error: 30.3254 - val_loss: 1068.0604 - val_mean_squared_error: 1068.0604
Epoch 52/100
81/81 [==============================] - 1s 8ms/step - loss: 27.4170 - mean_squared_error: 27.4170 - val_loss: 1322.2581 - val_mean_squared_error: 1322.2581
Epoch 53/100
81/81 [==============================] - 1s 8ms/step - loss: 25.0730 - mean_squared_error: 25.0730 - val_loss: 1207.5504 - val_mean_squared_error: 1207.5504
Epoch 54/100
81/81 [==============================] - 1s 8ms/step - loss: 29.2681 - mean_squared_error: 29.2681 - val_loss: 1110.3348 - val_mean_squared_error: 1110.3348
Epoch 55/100
81/81 [==============================] - 1s 8ms/step - loss: 27.2573 - mean_squared_error: 27.2573 - val_loss: 1195.2094 - val_mean_squared_error: 1195.2094
Epoch 56/100
81/81 [==============================] - 1s 8ms/step - loss: 26.5684 - mean_squared_error: 26.5684 - val_loss: 1145.2035 - val_mean_squared_error: 1145.2035
Epoch 57/100
81/81 [==============================] - 1s 8ms/step - loss: 24.1101 - mean_squared_error: 24.1101 - val_loss: 960.0526 - val_mean_squared_error: 960.0526
Epoch 58/100
81/81 [==============================] - 1s 7ms/step - loss: 31.9539 - mean_squared_error: 31.9539 - val_loss: 1183.9668 - val_mean_squared_error: 1183.9668
Epoch 59/100
81/81 [==============================] - 1s 7ms/step - loss: 27.3851 - mean_squared_error: 27.3851 - val_loss: 1158.4030 - val_mean_squared_error: 1158.4030
Epoch 60/100
81/81 [==============================] - 1s 8ms/step - loss: 27.7595 - mean_squared_error: 27.7595 - val_loss: 1308.2737 - val_mean_squared_error: 1308.2737
Epoch 61/100
81/81 [==============================] - 1s 8ms/step - loss: 22.8585 - mean_squared_error: 22.8585 - val_loss: 1182.7795 - val_mean_squared_error: 1182.7795
Epoch 62/100
81/81 [==============================] - 1s 8ms/step - loss: 28.2295 - mean_squared_error: 28.2295 - val_loss: 1155.2911 - val_mean_squared_error: 1155.2911
Epoch 63/100
81/81 [==============================] - 1s 8ms/step - loss: 29.3590 - mean_squared_error: 29.3590 - val_loss: 967.9388 - val_mean_squared_error: 967.9388
Epoch 64/100
81/81 [==============================] - 1s 8ms/step - loss: 23.8962 - mean_squared_error: 23.8962 - val_loss: 1063.1406 - val_mean_squared_error: 1063.1406
Epoch 65/100
81/81 [==============================] - 1s 8ms/step - loss: 25.5377 - mean_squared_error: 25.5377 - val_loss: 1222.8542 - val_mean_squared_error: 1222.8542
Epoch 66/100
81/81 [==============================] - 1s 8ms/step - loss: 25.9025 - mean_squared_error: 25.9025 - val_loss: 1067.4923 - val_mean_squared_error: 1067.4923
Epoch 67/100
81/81 [==============================] - 1s 8ms/step - loss: 27.5574 - mean_squared_error: 27.5574 - val_loss: 993.5450 - val_mean_squared_error: 993.5450
Epoch 68/100
81/81 [==============================] - 1s 8ms/step - loss: 27.0941 - mean_squared_error: 27.0941 - val_loss: 1121.9958 - val_mean_squared_error: 1121.9958
Epoch 69/100
81/81 [==============================] - 1s 8ms/step - loss: 21.6754 - mean_squared_error: 21.6754 - val_loss: 1152.4788 - val_mean_squared_error: 1152.4788
Epoch 70/100
81/81 [==============================] - 1s 9ms/step - loss: 27.4994 - mean_squared_error: 27.4994 - val_loss: 800.1244 - val_mean_squared_error: 800.1244
Epoch 71/100
81/81 [==============================] - 1s 10ms/step - loss: 24.9558 - mean_squared_error: 24.9558 - val_loss: 959.6923 - val_mean_squared_error: 959.6923
Epoch 72/100
81/81 [==============================] - 1s 8ms/step - loss: 26.0471 - mean_squared_error: 26.0471 - val_loss: 1192.9941 - val_mean_squared_error: 1192.9941
Epoch 73/100
81/81 [==============================] - 1s 8ms/step - loss: 26.6168 - mean_squared_error: 26.6168 - val_loss: 926.8901 - val_mean_squared_error: 926.8901
Epoch 74/100
81/81 [==============================] - 1s 8ms/step - loss: 25.8059 - mean_squared_error: 25.8059 - val_loss: 946.4683 - val_mean_squared_error: 946.4683
Epoch 75/100
81/81 [==============================] - 1s 8ms/step - loss: 24.7733 - mean_squared_error: 24.7733 - val_loss: 1001.5833 - val_mean_squared_error: 1001.5833
Epoch 76/100
81/81 [==============================] - 1s 8ms/step - loss: 25.2519 - mean_squared_error: 25.2519 - val_loss: 1045.1542 - val_mean_squared_error: 1045.1541
Epoch 77/100
81/81 [==============================] - 1s 8ms/step - loss: 22.3246 - mean_squared_error: 22.3246 - val_loss: 1106.0219 - val_mean_squared_error: 1106.0219
Epoch 78/100
81/81 [==============================] - 1s 8ms/step - loss: 23.7462 - mean_squared_error: 23.7462 - val_loss: 1151.8929 - val_mean_squared_error: 1151.8929
Epoch 79/100
81/81 [==============================] - 1s 8ms/step - loss: 22.8883 - mean_squared_error: 22.8883 - val_loss: 1127.8688 - val_mean_squared_error: 1127.8688
Epoch 80/100
81/81 [==============================] - 1s 8ms/step - loss: 24.5986 - mean_squared_error: 24.5986 - val_loss: 921.5747 - val_mean_squared_error: 921.5747
Epoch 81/100
81/81 [==============================] - 1s 8ms/step - loss: 25.1967 - mean_squared_error: 25.1967 - val_loss: 1052.8248 - val_mean_squared_error: 1052.8248
Epoch 82/100
81/81 [==============================] - 1s 8ms/step - loss: 26.5214 - mean_squared_error: 26.5214 - val_loss: 1038.0441 - val_mean_squared_error: 1038.0441
Epoch 83/100
81/81 [==============================] - 1s 8ms/step - loss: 22.5153 - mean_squared_error: 22.5153 - val_loss: 1091.1796 - val_mean_squared_error: 1091.1796
Epoch 84/100
81/81 [==============================] - 1s 8ms/step - loss: 24.6146 - mean_squared_error: 24.6146 - val_loss: 868.6038 - val_mean_squared_error: 868.6038
Epoch 85/100
81/81 [==============================] - 1s 8ms/step - loss: 27.5688 - mean_squared_error: 27.5688 - val_loss: 1117.9799 - val_mean_squared_error: 1117.9799
Epoch 86/100
81/81 [==============================] - 1s 8ms/step - loss: 20.5549 - mean_squared_error: 20.5549 - val_loss: 996.0103 - val_mean_squared_error: 996.0103
Epoch 87/100
81/81 [==============================] - 1s 8ms/step - loss: 20.8087 - mean_squared_error: 20.8087 - val_loss: 894.9278 - val_mean_squared_error: 894.9278
Epoch 88/100
81/81 [==============================] - 1s 8ms/step - loss: 21.1017 - mean_squared_error: 21.1017 - val_loss: 882.6274 - val_mean_squared_error: 882.6274
Epoch 89/100
81/81 [==============================] - 1s 8ms/step - loss: 22.9732 - mean_squared_error: 22.9732 - val_loss: 1019.8044 - val_mean_squared_error: 1019.8044
Epoch 90/100
81/81 [==============================] - 1s 8ms/step - loss: 23.1910 - mean_squared_error: 23.1910 - val_loss: 1063.9961 - val_mean_squared_error: 1063.9961
Epoch 91/100
81/81 [==============================] - 1s 8ms/step - loss: 23.0422 - mean_squared_error: 23.0422 - val_loss: 759.1998 - val_mean_squared_error: 759.1998
Epoch 92/100
81/81 [==============================] - 1s 8ms/step - loss: 23.2884 - mean_squared_error: 23.2884 - val_loss: 933.0878 - val_mean_squared_error: 933.0878
Epoch 93/100
81/81 [==============================] - 1s 8ms/step - loss: 22.2951 - mean_squared_error: 22.2951 - val_loss: 1040.7745 - val_mean_squared_error: 1040.7745
Epoch 94/100
81/81 [==============================] - 1s 8ms/step - loss: 23.6692 - mean_squared_error: 23.6692 - val_loss: 793.8560 - val_mean_squared_error: 793.8560
Epoch 95/100
81/81 [==============================] - 1s 8ms/step - loss: 21.5573 - mean_squared_error: 21.5573 - val_loss: 983.3247 - val_mean_squared_error: 983.3247
Epoch 96/100
81/81 [==============================] - 1s 8ms/step - loss: 23.4569 - mean_squared_error: 23.4569 - val_loss: 826.7446 - val_mean_squared_error: 826.7446
Epoch 97/100
81/81 [==============================] - 1s 7ms/step - loss: 23.0245 - mean_squared_error: 23.0245 - val_loss: 795.0195 - val_mean_squared_error: 795.0195
Epoch 98/100
81/81 [==============================] - 1s 8ms/step - loss: 21.2434 - mean_squared_error: 21.2434 - val_loss: 1019.3207 - val_mean_squared_error: 1019.3207
Epoch 99/100
81/81 [==============================] - 1s 7ms/step - loss: 21.3902 - mean_squared_error: 21.3902 - val_loss: 1031.2543 - val_mean_squared_error: 1031.2543
Epoch 100/100
81/81 [==============================] - 1s 7ms/step - loss: 20.1082 - mean_squared_error: 20.1082 - val_loss: 959.8607 - val_mean_squared_error: 959.8607


y_pred__test_ann = model_nn.predict(X_test)
y_pred__train_ann = model_nn.predict(X_train)

2022-12-09 17:28:55.241368: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.


plt.figure(figsize = (12,6), dpi = 400)
plt.plot(y_train_ann, label = 'Actual')
plt.plot(y_pred__train_ga_nn, label = 'GA-ANN-F')
plt.plot(y_pred__train_ann, label = 'ANN')

plt.legend()
#plt.savefig('Plot-GA-ANN_train.jpg', dpi = 800)


plt.figure(figsize = (12,6), dpi = 400)
plt.plot(y_test_ann, label = 'Actual')
plt.plot(y_pred__test_ga_nn, label = 'GA-ANN-F')
plt.plot(y_pred__test_ann, label = 'ANN')
plt.legend()
#plt.savefig('Plot-GA-ANN-F_test.jpg', dpi = 800)

<matplotlib.legend.Legend at 0x2c41ae340>


arr_nn_ga_f = list(y_pred__train_ga_nn) + list(y_pred__test_ga_nn)
arr_ann = list(y_pred__train_ann) + list(y_pred__test_ann)
actual = list(y_train) + list(y_test)


arr_nn_ga_f1 = []
for x in arr_nn_ga_f:
    arr_nn_ga_f1.append(float(x))
    
arr_ann1 = []
for x in arr_ann:
    arr_ann1.append(float(x))
    
actual1 = []
for x in actual:
    actual1.append(float(x))


temp = pd.DataFrame(date_arr)


temp['Actual'] = actual1


temp['ANN'] = arr_ann1
temp['GA_ANN_F'] = arr_nn_ga_f1


temp


plt.figure(figsize = (12,6), dpi = 600)

plt.plot( temp['DATE'],temp['Actual'], label = 'Actual')
plt.plot( temp['DATE'],temp['ANN'], label = 'ANN')
plt.plot( temp['DATE'],temp['GA_ANN_F'], label = 'GA-ANN-F')

plt.vlines(x = datetime(2018,8,18), ymin = 0, ymax= 240, color = 'red', linewidth = 2)
#plt.plot(predictions_ga_ann_test, label = 'ANN_GA',  marker = '*', alpha  =  0.5)
#plt.plot(predictions_ga_ann_test, label = 'LSTM_GA', alpha  =  0.5)
plt.legend()
plt.grid(linestyle = '-.', alpha = 0.5)
plt.ylim(0,160)
#plt.savefig('PlotGA_ANN_F.jpg', dpi = 800)
plt.show()


#temp.to_csv('temp1.csv')


def reduce_features(solution, features):
    
    selected_elements_indices = numpy.where(solution == 1)[0]
    reduced_features = features[:, selected_elements_indices]
    return reduced_features

def fitness_func(labels, predictions):
    mse = tf.keras.losses.MeanSquaredError()
    fitness = 1/(mse(labels, predictions).numpy() + 0.00001)
    return fitness

def cal_pop_fitness(pop, features, labels, train_indices, test_indices):
    accuracies = np.zeros(pop.shape[0])
    idx = 0
    for curr_solution in pop:
        reduced_features = reduce_features(curr_solution, features)
        train_data = reduced_features[train_indices, :]
        test_data = reduced_features[test_indices, :]
        train_labels = labels[train_indices]
        test_labels = labels[test_indices]
        print(test_data)
        print(train_data.shape)
        print(test_data.shape)
        model_lstm = Sequential()
        model_lstm.add(LSTM(128, input_shape= (train_data.shape[1],1), return_sequences=True))
        model_lstm.add(Dropout(0.2))
        model_lstm.add(BatchNormalization())  #normalizes activation outputs, same reason you want to normalize your input data.

        model_lstm.add(LSTM(128, return_sequences=True))
        model_lstm.add(Dropout(0.1))
        model_lstm.add(BatchNormalization())

        model_lstm.add(LSTM(128))
        model_lstm.add(Dropout(0.2))
        model_lstm.add(BatchNormalization())

        model_lstm.add(Dense(32, activation='relu'))
        model_lstm.add(Dropout(0.2))

        model_lstm.add(Dense(1, activation='relu'))

        opt = tf.keras.optimizers.Adam(lr=0.001, decay=1e-6)
        
        model_lstm.compile(optimizer=opt,  loss='mean_squared_error', metrics='mean_squared_error')
        history_nn = model_lstm.fit(train_data, train_labels, epochs = 30, validation_data = [test_data, test_labels])
        predictions = model_lstm.predict(test_data)
        accuracies[idx] = fitness_func(test_labels, predictions)
        idx = idx + 1
    return accuracies

def select_mating_pool(pop, fitness, num_parents):
    parents = numpy.empty((num_parents, pop_shape[1]))
    for parent_num in range(num_parents):
        max_fitness_idx = np.where(fitness == np.max(fitness))
        max_fitness_idx = max_fitness_idx[0][0]
        parents[parent_num, :] = pop[max_fitness_idx, :]
        fitness[max_fitness_idx] = -99999999999
    return parents

def crossover(parents, offspring_size):
    offspring = numpy.empty(offspring_size)
    # The point at which crossover takes place between two parents. Usually, it is at the center.
    crossover_point = numpy.uint8(offspring_size[1]/2)

    for k in range(offspring_size[0]):
        # Index of the first parent to mate.
        parent1_idx = k%parents.shape[0]
        # Index of the second parent to mate.
        parent2_idx = (k+1)%parents.shape[0]
        # The new offspring will have its first half of its genes taken from the first parent.
        offspring[k, 0:crossover_point] = parents[parent1_idx, 0:crossover_point]
        # The new offspring will have its second half of its genes taken from the second parent.
        offspring[k, crossover_point:] = parents[parent2_idx, crossover_point:]
    return offspring


def mutation(offspring_crossover, num_mutations = 2):
    mutation_idx = np.random.randint(low = 0, high = offspring_crossover.shape[1], size = num_mutations)
    for idx in range(offspring_crossover.shape[0]):
        offspring_crossover[idx, mutation_idx] = 1 - offspring_crossover[idx, mutation_idx]
    return offspring_crossover


num_generations = 20


import numpy


for generation in range(num_generations):
    print("Generation: ", generation)
    fitness = cal_pop_fitness(new_population, data_inputs, data_outputs, train_indices, test_indices)
    best_outputs.append(np.max(fitness))
    print("Best Result: ", best_outputs[-1])

    parents = select_mating_pool(new_population, fitness, num_parents_mating)
    offspring_crossover = crossover(parents, offspring_size= (pop_shape[0]-parents.shape[0], num_feature_elements))
    offspring_mutation = mutation(offspring_crossover, num_mutations= num_mutations)
    
    new_population[0:parents.shape[0], :] = parents
    new_population[parents.shape[0]:, :] = offspring_mutation
    fitness = cal_pop_fitness(new_population, data_inputs, data_outputs, train_indices, test_indices)
# Then return the index of that solution corresponding to the best fitness.
    best_match_idx = numpy.where(fitness == numpy.max(fitness))[0]
    best_match_idx = best_match_idx[0]

    best_solution = new_population[best_match_idx, :]
    best_solution_indices = numpy.where(best_solution == 1)[0]
    best_solution_num_elements = best_solution_indices.shape[0]
    best_solution_fitness = fitness[best_match_idx]

    print("best_match_idx : ", best_match_idx)
    print("best_solution : ", best_solution)
    print("Selected indices : ", best_solution_indices)
    print("Number of selected elements : ", best_solution_num_elements)
    print("Best solution fitness : ", best_solution_fitness)

Generation:  0
[[0.18225377 0.12548801 0.43662912 ... 0.         0.         0.        ]
 [0.19041704 0.15281651 0.44309554 ... 0.         0.         0.        ]
 [0.18030169 0.18600112 0.44272007 ... 0.         0.         0.        ]
 ...
 [0.07808341 0.07933631 0.74735086 ... 0.         0.         0.        ]
 [0.0743567  0.07306191 0.76512307 ... 0.         0.         0.        ]
 [0.07488909 0.08951478 0.76583229 ... 0.         0.         0.        ]]
(2577, 41)
(286, 41)
Metal device set to: Apple M1 Pro

2022-12-10 12:12:15.247000: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:305] Could not identify NUMA node of platform GPU ID 0, defaulting to 0. Your kernel may not have been built with NUMA support.
2022-12-10 12:12:15.247487: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:271] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 0 MB memory) -> physical PluggableDevice (device: 0, name: METAL, pci bus id: <undefined>)
2022-12-10 12:12:15.591995: W tensorflow/core/platform/profile_utils/cpu_utils.cc:128] Failed to get CPU frequency: 0 Hz

Epoch 1/30

2022-12-10 12:12:16.613049: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.
2022-12-10 12:12:17.025714: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.
2022-12-10 12:12:17.145853: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.
2022-12-10 12:12:17.227558: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.
2022-12-10 12:12:17.408250: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.
2022-12-10 12:12:17.595232: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.
2022-12-10 12:12:17.741336: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.

81/81 [==============================] - ETA: 0s - loss: 4714.0571 - mean_squared_error: 4714.0571

2022-12-10 12:12:21.678683: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.
2022-12-10 12:12:21.808935: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.
2022-12-10 12:12:21.871977: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.
2022-12-10 12:12:21.943319: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.

81/81 [==============================] - 7s 55ms/step - loss: 4714.0571 - mean_squared_error: 4714.0571 - val_loss: 375.3421 - val_mean_squared_error: 375.3421
Epoch 2/30
81/81 [==============================] - 3s 41ms/step - loss: 2280.6301 - mean_squared_error: 2280.6301 - val_loss: 79.6511 - val_mean_squared_error: 79.6511
Epoch 3/30
81/81 [==============================] - 3s 42ms/step - loss: 1235.7565 - mean_squared_error: 1235.7565 - val_loss: 317.3334 - val_mean_squared_error: 317.3334
Epoch 4/30
81/81 [==============================] - 3s 42ms/step - loss: 1069.3400 - mean_squared_error: 1069.3400 - val_loss: 39.1712 - val_mean_squared_error: 39.1712
Epoch 5/30
81/81 [==============================] - 3s 42ms/step - loss: 838.7263 - mean_squared_error: 838.7263 - val_loss: 872.5482 - val_mean_squared_error: 872.5482
Epoch 6/30
81/81 [==============================] - 3s 42ms/step - loss: 624.9553 - mean_squared_error: 624.9553 - val_loss: 96.6863 - val_mean_squared_error: 96.6863
Epoch 7/30
81/81 [==============================] - 3s 42ms/step - loss: 572.8158 - mean_squared_error: 572.8158 - val_loss: 118.2364 - val_mean_squared_error: 118.2364
Epoch 8/30
81/81 [==============================] - 3s 42ms/step - loss: 508.6587 - mean_squared_error: 508.6587 - val_loss: 63.4274 - val_mean_squared_error: 63.4274
Epoch 9/30
81/81 [==============================] - 3s 42ms/step - loss: 500.4796 - mean_squared_error: 500.4796 - val_loss: 87.6345 - val_mean_squared_error: 87.6345
Epoch 10/30
81/81 [==============================] - 3s 41ms/step - loss: 514.3248 - mean_squared_error: 514.3248 - val_loss: 140.3179 - val_mean_squared_error: 140.3179
Epoch 11/30
81/81 [==============================] - 3s 41ms/step - loss: 460.7059 - mean_squared_error: 460.7059 - val_loss: 66.2669 - val_mean_squared_error: 66.2669
Epoch 12/30
81/81 [==============================] - 3s 42ms/step - loss: 465.5082 - mean_squared_error: 465.5082 - val_loss: 21.7861 - val_mean_squared_error: 21.7861
Epoch 13/30
81/81 [==============================] - 3s 42ms/step - loss: 481.7058 - mean_squared_error: 481.7058 - val_loss: 130.1794 - val_mean_squared_error: 130.1794
Epoch 14/30
81/81 [==============================] - 3s 42ms/step - loss: 478.5070 - mean_squared_error: 478.5070 - val_loss: 32.4945 - val_mean_squared_error: 32.4945
Epoch 15/30
81/81 [==============================] - 3s 42ms/step - loss: 492.7027 - mean_squared_error: 492.7027 - val_loss: 1079.7848 - val_mean_squared_error: 1079.7848
Epoch 16/30
81/81 [==============================] - 3s 42ms/step - loss: 314.0490 - mean_squared_error: 314.0490 - val_loss: 57.7133 - val_mean_squared_error: 57.7133
Epoch 17/30
81/81 [==============================] - 3s 42ms/step - loss: 285.6759 - mean_squared_error: 285.6759 - val_loss: 41.2296 - val_mean_squared_error: 41.2296
Epoch 18/30
81/81 [==============================] - 3s 42ms/step - loss: 271.2457 - mean_squared_error: 271.2457 - val_loss: 57.8656 - val_mean_squared_error: 57.8656
Epoch 19/30
81/81 [==============================] - 3s 42ms/step - loss: 269.8030 - mean_squared_error: 269.8030 - val_loss: 26.7355 - val_mean_squared_error: 26.7355
Epoch 20/30
81/81 [==============================] - 3s 42ms/step - loss: 239.2381 - mean_squared_error: 239.2381 - val_loss: 94.7504 - val_mean_squared_error: 94.7504
Epoch 21/30
81/81 [==============================] - 4s 43ms/step - loss: 258.2213 - mean_squared_error: 258.2213 - val_loss: 64.6303 - val_mean_squared_error: 64.6303
Epoch 22/30
81/81 [==============================] - 3s 43ms/step - loss: 271.4872 - mean_squared_error: 271.4872 - val_loss: 47.4784 - val_mean_squared_error: 47.4784
Epoch 23/30
81/81 [==============================] - 3s 42ms/step - loss: 235.2861 - mean_squared_error: 235.2861 - val_loss: 107.2364 - val_mean_squared_error: 107.2364
Epoch 24/30
81/81 [==============================] - 3s 42ms/step - loss: 281.7242 - mean_squared_error: 281.7242 - val_loss: 80.6768 - val_mean_squared_error: 80.6768
Epoch 25/30
81/81 [==============================] - 4s 43ms/step - loss: 246.9300 - mean_squared_error: 246.9300 - val_loss: 155.3753 - val_mean_squared_error: 155.3753
Epoch 26/30
81/81 [==============================] - 4s 44ms/step - loss: 239.4642 - mean_squared_error: 239.4642 - val_loss: 69.5593 - val_mean_squared_error: 69.5593
Epoch 27/30
81/81 [==============================] - 3s 43ms/step - loss: 259.5128 - mean_squared_error: 259.5128 - val_loss: 64.6172 - val_mean_squared_error: 64.6172
Epoch 28/30
81/81 [==============================] - 3s 43ms/step - loss: 234.5875 - mean_squared_error: 234.5875 - val_loss: 56.0815 - val_mean_squared_error: 56.0815
Epoch 29/30
81/81 [==============================] - 3s 43ms/step - loss: 234.0160 - mean_squared_error: 234.0160 - val_loss: 143.1393 - val_mean_squared_error: 143.1393
Epoch 30/30
81/81 [==============================] - 4s 43ms/step - loss: 231.5211 - mean_squared_error: 231.5211 - val_loss: 151.6738 - val_mean_squared_error: 151.6738

2022-12-10 12:14:01.644428: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.
2022-12-10 12:14:01.760846: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.
2022-12-10 12:14:01.823043: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.
2022-12-10 12:14:01.892490: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.

[[0.6994447  0.60500742 0.12548801 ... 0.         0.         0.        ]
 [0.6737802  0.60709947 0.15281651 ... 0.         0.         0.        ]
 [0.65978331 0.61047375 0.18600112 ... 0.         0.         0.        ]
 ...
 [0.99012252 0.64428398 0.07933631 ... 0.         0.         0.        ]
 [1.         0.6460386  0.07306191 ... 0.         0.         0.        ]
 [0.99225423 0.66290997 0.08951478 ... 0.         0.         0.        ]]
(2577, 44)
(286, 44)
Epoch 1/30

2022-12-10 12:14:03.413270: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.
2022-12-10 12:14:03.840739: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.
2022-12-10 12:14:03.929847: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.
2022-12-10 12:14:04.021430: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.
2022-12-10 12:14:04.168059: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.
2022-12-10 12:14:04.325669: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.

[1,
 2,
 3,
 4,
 6,
 7,
 12,
 14,
 15,
 17,
 22,
 25,
 28,
 31,
 33,
 34,
 35,
 36,
 37,
 38,
 41,
 45,
 49,
 50,
 58,
 59,
 60,
 61,
 62,
 67,
 68,
 69,
 72,
 73,
 78,
 79]


reduced_data_inputs1 = data_inputs[:, [1,  2,  3,  4,  6,  7, 12, 14, 15, 17, 22, 25, 28, 31, 33, 34, 35, 36, 37, 38, 41, 45, 49, 50,
 58, 59, 60, 61, 62, 67, 68, 69, 72, 73, 78, 79]]


train_data_ga_ann_f = reduced_data_inputs1[train_indices, :]
test_data_ga_ann_f = reduced_data_inputs1[test_indices, :]


y_train_ann = data_outputs[train_indices]

y_test_ann = data_outputs[test_indices]


model_lstm = Sequential()
model_lstm.add(LSTM(128, input_shape= (X_train.shape[1],1), return_sequences=True))
model_lstm.add(Dropout(0.2))
model_lstm.add(BatchNormalization())  #normalizes activation outputs, same reason you want to normalize your input data.
model_lstm.add(LSTM(128, return_sequences=True))
model_lstm.add(Dropout(0.1))
model_lstm.add(BatchNormalization())
model_lstm.add(LSTM(128))
model_lstm.add(Dropout(0.2))
model_lstm.add(BatchNormalization())
model_lstm.add(Dense(32, activation='relu'))
model_lstm.add(Dropout(0.2))
model_lstm.add(Dense(1, activation='relu'))
opt = tf.keras.optimizers.Adam(lr=0.001, decay=1e-6)
model_lstm.compile(optimizer=opt,  loss='mean_squared_error', metrics='mean_squared_error')
history_lstm = model_lstm.fit(X_train, y_train, epochs = 100, validation_data = [X_test, y_test])

2022-12-10 17:33:08.133598: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:305] Could not identify NUMA node of platform GPU ID 0, defaulting to 0. Your kernel may not have been built with NUMA support.
2022-12-10 17:33:08.133995: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:271] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 0 MB memory) -> physical PluggableDevice (device: 0, name: METAL, pci bus id: <undefined>)

Metal device set to: Apple M1 Pro

2022-12-10 17:33:08.481782: W tensorflow/core/platform/profile_utils/cpu_utils.cc:128] Failed to get CPU frequency: 0 Hz

Epoch 1/100

2022-12-10 17:33:09.405663: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.
2022-12-10 17:33:09.811279: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.
2022-12-10 17:33:10.091436: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.
2022-12-10 17:33:10.210101: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.
2022-12-10 17:33:10.424218: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.
2022-12-10 17:33:10.708692: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.

 1/81 [..............................] - ETA: 3:30 - loss: 5699.7642 - mean_squared_error: 5699.7642

2022-12-10 17:33:10.925137: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.

81/81 [==============================] - ETA: 0s - loss: 4522.9033 - mean_squared_error: 4522.9033

2022-12-10 17:33:16.782585: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.
2022-12-10 17:33:16.909764: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.
2022-12-10 17:33:17.008896: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.
2022-12-10 17:33:17.114728: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.

81/81 [==============================] - 9s 82ms/step - loss: 4522.9033 - mean_squared_error: 4522.9033 - val_loss: 2601.5344 - val_mean_squared_error: 2601.5344
Epoch 2/100
81/81 [==============================] - 5s 61ms/step - loss: 1987.4360 - mean_squared_error: 1987.4360 - val_loss: 2557.0552 - val_mean_squared_error: 2557.0552
Epoch 3/100
81/81 [==============================] - 5s 62ms/step - loss: 555.0073 - mean_squared_error: 555.0073 - val_loss: 1976.2987 - val_mean_squared_error: 1976.2987
Epoch 4/100
81/81 [==============================] - 5s 62ms/step - loss: 143.4627 - mean_squared_error: 143.4627 - val_loss: 1591.7965 - val_mean_squared_error: 1591.7965
Epoch 5/100
81/81 [==============================] - 5s 62ms/step - loss: 97.0586 - mean_squared_error: 97.0586 - val_loss: 1016.9556 - val_mean_squared_error: 1016.9556
Epoch 6/100
81/81 [==============================] - 5s 62ms/step - loss: 74.3813 - mean_squared_error: 74.3813 - val_loss: 547.1500 - val_mean_squared_error: 547.1500
Epoch 7/100
81/81 [==============================] - 5s 62ms/step - loss: 63.6429 - mean_squared_error: 63.6429 - val_loss: 820.0898 - val_mean_squared_error: 820.0898
Epoch 8/100
81/81 [==============================] - 5s 62ms/step - loss: 82.5693 - mean_squared_error: 82.5693 - val_loss: 340.4031 - val_mean_squared_error: 340.4031
Epoch 9/100
81/81 [==============================] - 5s 62ms/step - loss: 56.4685 - mean_squared_error: 56.4685 - val_loss: 456.5211 - val_mean_squared_error: 456.5211
Epoch 10/100
81/81 [==============================] - 5s 64ms/step - loss: 55.4697 - mean_squared_error: 55.4697 - val_loss: 476.0818 - val_mean_squared_error: 476.0818
Epoch 11/100
81/81 [==============================] - 5s 64ms/step - loss: 49.3165 - mean_squared_error: 49.3165 - val_loss: 465.6936 - val_mean_squared_error: 465.6936
Epoch 12/100
81/81 [==============================] - 5s 65ms/step - loss: 48.8402 - mean_squared_error: 48.8402 - val_loss: 214.9186 - val_mean_squared_error: 214.9186
Epoch 13/100
81/81 [==============================] - 5s 64ms/step - loss: 51.4876 - mean_squared_error: 51.4876 - val_loss: 345.3933 - val_mean_squared_error: 345.3933
Epoch 14/100
81/81 [==============================] - 5s 63ms/step - loss: 56.1113 - mean_squared_error: 56.1113 - val_loss: 123.8209 - val_mean_squared_error: 123.8209
Epoch 15/100
81/81 [==============================] - 5s 65ms/step - loss: 50.8022 - mean_squared_error: 50.8022 - val_loss: 37.4062 - val_mean_squared_error: 37.4062
Epoch 16/100
81/81 [==============================] - 5s 64ms/step - loss: 40.9378 - mean_squared_error: 40.9378 - val_loss: 235.5872 - val_mean_squared_error: 235.5872
Epoch 17/100
81/81 [==============================] - 5s 64ms/step - loss: 46.0199 - mean_squared_error: 46.0199 - val_loss: 229.9056 - val_mean_squared_error: 229.9056
Epoch 18/100
81/81 [==============================] - 5s 63ms/step - loss: 37.4917 - mean_squared_error: 37.4917 - val_loss: 517.0174 - val_mean_squared_error: 517.0174
Epoch 19/100
81/81 [==============================] - 5s 64ms/step - loss: 37.4263 - mean_squared_error: 37.4263 - val_loss: 529.3101 - val_mean_squared_error: 529.3101
Epoch 20/100
81/81 [==============================] - 5s 65ms/step - loss: 37.6654 - mean_squared_error: 37.6654 - val_loss: 128.6450 - val_mean_squared_error: 128.6450
Epoch 21/100
81/81 [==============================] - 5s 66ms/step - loss: 36.2736 - mean_squared_error: 36.2736 - val_loss: 343.9818 - val_mean_squared_error: 343.9818
Epoch 22/100
81/81 [==============================] - 5s 66ms/step - loss: 33.4098 - mean_squared_error: 33.4098 - val_loss: 246.2173 - val_mean_squared_error: 246.2173
Epoch 23/100
81/81 [==============================] - 5s 65ms/step - loss: 33.5554 - mean_squared_error: 33.5554 - val_loss: 251.4927 - val_mean_squared_error: 251.4927
Epoch 24/100
81/81 [==============================] - 5s 64ms/step - loss: 32.5110 - mean_squared_error: 32.5110 - val_loss: 123.1326 - val_mean_squared_error: 123.1326
Epoch 25/100
81/81 [==============================] - 5s 64ms/step - loss: 35.3434 - mean_squared_error: 35.3434 - val_loss: 424.6902 - val_mean_squared_error: 424.6902
Epoch 26/100
81/81 [==============================] - 5s 65ms/step - loss: 35.1958 - mean_squared_error: 35.1958 - val_loss: 201.4658 - val_mean_squared_error: 201.4658
Epoch 27/100
81/81 [==============================] - 5s 64ms/step - loss: 32.8944 - mean_squared_error: 32.8944 - val_loss: 59.8623 - val_mean_squared_error: 59.8623
Epoch 28/100
81/81 [==============================] - 5s 65ms/step - loss: 33.7596 - mean_squared_error: 33.7596 - val_loss: 156.6700 - val_mean_squared_error: 156.6700
Epoch 29/100
81/81 [==============================] - 5s 63ms/step - loss: 41.3333 - mean_squared_error: 41.3333 - val_loss: 40.6560 - val_mean_squared_error: 40.6560
Epoch 30/100
81/81 [==============================] - 5s 64ms/step - loss: 26.8320 - mean_squared_error: 26.8320 - val_loss: 431.3963 - val_mean_squared_error: 431.3963
Epoch 31/100
81/81 [==============================] - 5s 65ms/step - loss: 43.5702 - mean_squared_error: 43.5702 - val_loss: 52.5264 - val_mean_squared_error: 52.5264
Epoch 32/100
81/81 [==============================] - 5s 66ms/step - loss: 30.8763 - mean_squared_error: 30.8763 - val_loss: 67.0091 - val_mean_squared_error: 67.0091
Epoch 33/100
81/81 [==============================] - 5s 66ms/step - loss: 31.7585 - mean_squared_error: 31.7585 - val_loss: 211.8969 - val_mean_squared_error: 211.8969
Epoch 34/100
81/81 [==============================] - 5s 66ms/step - loss: 32.0688 - mean_squared_error: 32.0688 - val_loss: 431.5067 - val_mean_squared_error: 431.5067
Epoch 35/100
81/81 [==============================] - 5s 67ms/step - loss: 27.7793 - mean_squared_error: 27.7793 - val_loss: 144.9303 - val_mean_squared_error: 144.9303
Epoch 36/100
81/81 [==============================] - 5s 65ms/step - loss: 29.8616 - mean_squared_error: 29.8616 - val_loss: 323.2311 - val_mean_squared_error: 323.2311
Epoch 37/100
81/81 [==============================] - 5s 65ms/step - loss: 29.4070 - mean_squared_error: 29.4070 - val_loss: 107.4282 - val_mean_squared_error: 107.4282
Epoch 38/100
81/81 [==============================] - 5s 65ms/step - loss: 27.3842 - mean_squared_error: 27.3842 - val_loss: 23.1213 - val_mean_squared_error: 23.1213
Epoch 39/100
81/81 [==============================] - 5s 65ms/step - loss: 32.7820 - mean_squared_error: 32.7820 - val_loss: 35.9743 - val_mean_squared_error: 35.9743
Epoch 40/100
81/81 [==============================] - 5s 66ms/step - loss: 23.3656 - mean_squared_error: 23.3656 - val_loss: 119.1868 - val_mean_squared_error: 119.1868
Epoch 41/100
81/81 [==============================] - 5s 65ms/step - loss: 51.8182 - mean_squared_error: 51.8182 - val_loss: 47.6383 - val_mean_squared_error: 47.6383
Epoch 42/100
81/81 [==============================] - 5s 65ms/step - loss: 36.7329 - mean_squared_error: 36.7329 - val_loss: 371.7837 - val_mean_squared_error: 371.7837
Epoch 43/100
81/81 [==============================] - 5s 64ms/step - loss: 28.8648 - mean_squared_error: 28.8648 - val_loss: 337.8975 - val_mean_squared_error: 337.8975
Epoch 44/100
81/81 [==============================] - 5s 65ms/step - loss: 27.5720 - mean_squared_error: 27.5720 - val_loss: 997.7277 - val_mean_squared_error: 997.7277
Epoch 45/100
81/81 [==============================] - 5s 65ms/step - loss: 25.0224 - mean_squared_error: 25.0224 - val_loss: 292.9283 - val_mean_squared_error: 292.9283
Epoch 46/100
81/81 [==============================] - 5s 65ms/step - loss: 24.3668 - mean_squared_error: 24.3668 - val_loss: 200.2119 - val_mean_squared_error: 200.2119
Epoch 47/100
81/81 [==============================] - 5s 65ms/step - loss: 22.8118 - mean_squared_error: 22.8118 - val_loss: 270.6850 - val_mean_squared_error: 270.6850
Epoch 48/100
81/81 [==============================] - 5s 65ms/step - loss: 23.0620 - mean_squared_error: 23.0620 - val_loss: 66.5645 - val_mean_squared_error: 66.5645
Epoch 49/100
81/81 [==============================] - 5s 66ms/step - loss: 22.0283 - mean_squared_error: 22.0283 - val_loss: 177.9130 - val_mean_squared_error: 177.9130
Epoch 50/100
81/81 [==============================] - 5s 65ms/step - loss: 24.1213 - mean_squared_error: 24.1213 - val_loss: 44.8422 - val_mean_squared_error: 44.8422
Epoch 51/100
81/81 [==============================] - 5s 64ms/step - loss: 26.2883 - mean_squared_error: 26.2883 - val_loss: 68.2217 - val_mean_squared_error: 68.2217
Epoch 52/100
81/81 [==============================] - 5s 65ms/step - loss: 20.6541 - mean_squared_error: 20.6541 - val_loss: 92.9119 - val_mean_squared_error: 92.9119
Epoch 53/100
81/81 [==============================] - 5s 64ms/step - loss: 19.4611 - mean_squared_error: 19.4611 - val_loss: 73.2735 - val_mean_squared_error: 73.2735
Epoch 54/100
81/81 [==============================] - 5s 65ms/step - loss: 20.0033 - mean_squared_error: 20.0033 - val_loss: 39.4270 - val_mean_squared_error: 39.4270
Epoch 55/100
81/81 [==============================] - 5s 65ms/step - loss: 18.9538 - mean_squared_error: 18.9538 - val_loss: 106.9972 - val_mean_squared_error: 106.9972
Epoch 56/100
81/81 [==============================] - 5s 65ms/step - loss: 20.6401 - mean_squared_error: 20.6401 - val_loss: 83.2301 - val_mean_squared_error: 83.2301
Epoch 57/100
81/81 [==============================] - 5s 64ms/step - loss: 22.0079 - mean_squared_error: 22.0079 - val_loss: 16.7783 - val_mean_squared_error: 16.7783
Epoch 58/100
81/81 [==============================] - 5s 65ms/step - loss: 18.0890 - mean_squared_error: 18.0890 - val_loss: 17.3304 - val_mean_squared_error: 17.3304
Epoch 59/100
81/81 [==============================] - 5s 65ms/step - loss: 20.6884 - mean_squared_error: 20.6884 - val_loss: 20.6213 - val_mean_squared_error: 20.6213
Epoch 60/100
81/81 [==============================] - 5s 65ms/step - loss: 18.2414 - mean_squared_error: 18.2414 - val_loss: 25.8314 - val_mean_squared_error: 25.8314
Epoch 61/100
81/81 [==============================] - 5s 65ms/step - loss: 18.0224 - mean_squared_error: 18.0224 - val_loss: 51.5101 - val_mean_squared_error: 51.5101
Epoch 62/100
81/81 [==============================] - 5s 66ms/step - loss: 17.1940 - mean_squared_error: 17.1940 - val_loss: 91.6027 - val_mean_squared_error: 91.6027
Epoch 63/100
81/81 [==============================] - 5s 65ms/step - loss: 19.4903 - mean_squared_error: 19.4903 - val_loss: 113.7526 - val_mean_squared_error: 113.7526
Epoch 64/100
81/81 [==============================] - 5s 65ms/step - loss: 17.7891 - mean_squared_error: 17.7891 - val_loss: 69.2880 - val_mean_squared_error: 69.2880
Epoch 65/100
81/81 [==============================] - 5s 65ms/step - loss: 17.6011 - mean_squared_error: 17.6011 - val_loss: 40.1573 - val_mean_squared_error: 40.1573
Epoch 66/100
81/81 [==============================] - 5s 65ms/step - loss: 16.7599 - mean_squared_error: 16.7599 - val_loss: 19.2697 - val_mean_squared_error: 19.2697
Epoch 67/100
81/81 [==============================] - 5s 65ms/step - loss: 16.7823 - mean_squared_error: 16.7823 - val_loss: 104.2594 - val_mean_squared_error: 104.2594
Epoch 68/100
81/81 [==============================] - 5s 65ms/step - loss: 16.0880 - mean_squared_error: 16.0880 - val_loss: 154.9836 - val_mean_squared_error: 154.9836
Epoch 69/100
81/81 [==============================] - 5s 65ms/step - loss: 18.1699 - mean_squared_error: 18.1699 - val_loss: 55.1115 - val_mean_squared_error: 55.1115
Epoch 70/100
81/81 [==============================] - 5s 66ms/step - loss: 17.9095 - mean_squared_error: 17.9095 - val_loss: 146.3438 - val_mean_squared_error: 146.3438
Epoch 71/100
81/81 [==============================] - 5s 65ms/step - loss: 20.4260 - mean_squared_error: 20.4260 - val_loss: 31.6194 - val_mean_squared_error: 31.6194
Epoch 72/100
81/81 [==============================] - 5s 66ms/step - loss: 17.4267 - mean_squared_error: 17.4267 - val_loss: 40.5962 - val_mean_squared_error: 40.5962
Epoch 73/100
81/81 [==============================] - 5s 66ms/step - loss: 16.4030 - mean_squared_error: 16.4030 - val_loss: 42.2971 - val_mean_squared_error: 42.2971
Epoch 74/100
81/81 [==============================] - 5s 66ms/step - loss: 17.4273 - mean_squared_error: 17.4273 - val_loss: 48.7428 - val_mean_squared_error: 48.7428
Epoch 75/100
81/81 [==============================] - 5s 66ms/step - loss: 16.6753 - mean_squared_error: 16.6753 - val_loss: 100.0497 - val_mean_squared_error: 100.0497
Epoch 76/100
81/81 [==============================] - 5s 66ms/step - loss: 16.8725 - mean_squared_error: 16.8725 - val_loss: 108.2646 - val_mean_squared_error: 108.2646
Epoch 77/100
81/81 [==============================] - 5s 66ms/step - loss: 17.1410 - mean_squared_error: 17.1410 - val_loss: 87.6227 - val_mean_squared_error: 87.6227
Epoch 78/100
81/81 [==============================] - 5s 65ms/step - loss: 17.3236 - mean_squared_error: 17.3236 - val_loss: 31.5636 - val_mean_squared_error: 31.5636
Epoch 79/100
81/81 [==============================] - 5s 65ms/step - loss: 18.2951 - mean_squared_error: 18.2951 - val_loss: 18.3175 - val_mean_squared_error: 18.3175
Epoch 80/100
81/81 [==============================] - 5s 68ms/step - loss: 17.2014 - mean_squared_error: 17.2014 - val_loss: 55.1481 - val_mean_squared_error: 55.1481
Epoch 81/100
81/81 [==============================] - 5s 66ms/step - loss: 15.4642 - mean_squared_error: 15.4642 - val_loss: 25.0469 - val_mean_squared_error: 25.0469
Epoch 82/100
81/81 [==============================] - 5s 66ms/step - loss: 15.1718 - mean_squared_error: 15.1718 - val_loss: 58.7281 - val_mean_squared_error: 58.7281
Epoch 83/100
81/81 [==============================] - 5s 66ms/step - loss: 15.1982 - mean_squared_error: 15.1982 - val_loss: 56.5707 - val_mean_squared_error: 56.5707
Epoch 84/100
81/81 [==============================] - 5s 66ms/step - loss: 15.1385 - mean_squared_error: 15.1385 - val_loss: 34.3927 - val_mean_squared_error: 34.3927
Epoch 85/100
81/81 [==============================] - 5s 67ms/step - loss: 14.1924 - mean_squared_error: 14.1924 - val_loss: 63.0380 - val_mean_squared_error: 63.0380
Epoch 86/100
81/81 [==============================] - 5s 66ms/step - loss: 14.8381 - mean_squared_error: 14.8381 - val_loss: 47.1645 - val_mean_squared_error: 47.1645
Epoch 87/100
81/81 [==============================] - 5s 66ms/step - loss: 15.1222 - mean_squared_error: 15.1222 - val_loss: 31.3233 - val_mean_squared_error: 31.3233
Epoch 88/100
81/81 [==============================] - 5s 66ms/step - loss: 15.2172 - mean_squared_error: 15.2172 - val_loss: 80.5635 - val_mean_squared_error: 80.5635
Epoch 89/100
81/81 [==============================] - 5s 66ms/step - loss: 13.3420 - mean_squared_error: 13.3420 - val_loss: 85.0002 - val_mean_squared_error: 85.0002
Epoch 90/100
81/81 [==============================] - 5s 67ms/step - loss: 12.6373 - mean_squared_error: 12.6373 - val_loss: 64.8516 - val_mean_squared_error: 64.8516
Epoch 91/100
81/81 [==============================] - 5s 66ms/step - loss: 13.3615 - mean_squared_error: 13.3615 - val_loss: 38.3302 - val_mean_squared_error: 38.3302
Epoch 92/100
81/81 [==============================] - 5s 66ms/step - loss: 13.1553 - mean_squared_error: 13.1553 - val_loss: 28.2080 - val_mean_squared_error: 28.2080
Epoch 93/100
81/81 [==============================] - 5s 66ms/step - loss: 15.6029 - mean_squared_error: 15.6029 - val_loss: 20.4873 - val_mean_squared_error: 20.4873
Epoch 94/100
81/81 [==============================] - 5s 66ms/step - loss: 11.9100 - mean_squared_error: 11.9100 - val_loss: 24.2193 - val_mean_squared_error: 24.2193
Epoch 95/100
81/81 [==============================] - 5s 66ms/step - loss: 12.2501 - mean_squared_error: 12.2501 - val_loss: 28.4298 - val_mean_squared_error: 28.4298
Epoch 96/100
81/81 [==============================] - 5s 66ms/step - loss: 14.2114 - mean_squared_error: 14.2114 - val_loss: 70.1251 - val_mean_squared_error: 70.1251
Epoch 97/100
81/81 [==============================] - 5s 65ms/step - loss: 14.1764 - mean_squared_error: 14.1764 - val_loss: 31.5748 - val_mean_squared_error: 31.5748
Epoch 98/100
81/81 [==============================] - 5s 65ms/step - loss: 14.3958 - mean_squared_error: 14.3958 - val_loss: 163.4696 - val_mean_squared_error: 163.4696
Epoch 99/100
81/81 [==============================] - 5s 65ms/step - loss: 12.4203 - mean_squared_error: 12.4203 - val_loss: 25.9320 - val_mean_squared_error: 25.9320
Epoch 100/100
81/81 [==============================] - 5s 65ms/step - loss: 13.5944 - mean_squared_error: 13.5944 - val_loss: 57.4852 - val_mean_squared_error: 57.4852


y_pred__test_lstm = model_lstm.predict(X_test)
y_pred__train_lstm = model_lstm.predict(X_train)

2022-12-10 17:42:43.954072: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.
2022-12-10 17:42:44.080087: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.
2022-12-10 17:42:44.185588: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.
2022-12-10 17:42:44.295754: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.


model_lstm = Sequential()
model_lstm.add(LSTM(128, input_shape= (train_data_ga_ann_f.shape[1],1), return_sequences=True))
model_lstm.add(Dropout(0.2))
model_lstm.add(BatchNormalization())  #normalizes activation outputs, same reason you want to normalize your input data.
model_lstm.add(LSTM(128, return_sequences=True))
model_lstm.add(Dropout(0.1))
model_lstm.add(BatchNormalization())
model_lstm.add(LSTM(128))
model_lstm.add(Dropout(0.2))
model_lstm.add(BatchNormalization())
model_lstm.add(Dense(32, activation='relu'))
model_lstm.add(Dropout(0.2))
model_lstm.add(Dense(1, activation='relu'))
opt = tf.keras.optimizers.Adam(lr=0.001, decay=1e-6)
model_lstm.compile(optimizer=opt,  loss='mean_squared_error', metrics='mean_squared_error')
history_lstm = model_lstm.fit(train_data_ga_ann_f, y_train_ann, epochs = 100, validation_data = [test_data_ga_ann_f, y_test_ann])

Epoch 1/100

2022-12-10 17:43:35.660292: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.
2022-12-10 17:43:36.094481: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.
2022-12-10 17:43:36.178367: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.
2022-12-10 17:43:36.263728: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.
2022-12-10 17:43:36.408387: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.
2022-12-10 17:43:36.564024: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.

 2/81 [..............................] - ETA: 5s - loss: 6668.0312 - mean_squared_error: 6668.0312

2022-12-10 17:43:36.716584: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.

81/81 [==============================] - ETA: 0s - loss: 4280.5874 - mean_squared_error: 4280.5874

2022-12-10 17:43:40.702041: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.
2022-12-10 17:43:40.850117: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.
2022-12-10 17:43:40.917395: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.
2022-12-10 17:43:41.001380: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.

81/81 [==============================] - 7s 57ms/step - loss: 4280.5874 - mean_squared_error: 4280.5874 - val_loss: 1154.5619 - val_mean_squared_error: 1154.5619
Epoch 2/100
81/81 [==============================] - 3s 43ms/step - loss: 1426.3562 - mean_squared_error: 1426.3562 - val_loss: 22.4790 - val_mean_squared_error: 22.4790
Epoch 3/100
81/81 [==============================] - 3s 41ms/step - loss: 583.6574 - mean_squared_error: 583.6574 - val_loss: 633.1177 - val_mean_squared_error: 633.1177
Epoch 4/100
81/81 [==============================] - 3s 41ms/step - loss: 628.4687 - mean_squared_error: 628.4687 - val_loss: 90.9474 - val_mean_squared_error: 90.9474
Epoch 5/100
81/81 [==============================] - 3s 40ms/step - loss: 278.8212 - mean_squared_error: 278.8212 - val_loss: 145.7620 - val_mean_squared_error: 145.7620
Epoch 6/100
81/81 [==============================] - 3s 40ms/step - loss: 191.0997 - mean_squared_error: 191.0997 - val_loss: 371.8455 - val_mean_squared_error: 371.8455
Epoch 7/100
81/81 [==============================] - 3s 40ms/step - loss: 172.7908 - mean_squared_error: 172.7908 - val_loss: 269.6066 - val_mean_squared_error: 269.6066
Epoch 8/100
81/81 [==============================] - 3s 40ms/step - loss: 126.8157 - mean_squared_error: 126.8157 - val_loss: 197.5990 - val_mean_squared_error: 197.5990
Epoch 9/100
81/81 [==============================] - 3s 40ms/step - loss: 106.8475 - mean_squared_error: 106.8475 - val_loss: 160.4723 - val_mean_squared_error: 160.4723
Epoch 10/100
81/81 [==============================] - 3s 40ms/step - loss: 97.3431 - mean_squared_error: 97.3431 - val_loss: 32.4610 - val_mean_squared_error: 32.4610
Epoch 11/100
81/81 [==============================] - 3s 40ms/step - loss: 93.9596 - mean_squared_error: 93.9596 - val_loss: 66.7283 - val_mean_squared_error: 66.7283
Epoch 12/100
81/81 [==============================] - 3s 40ms/step - loss: 85.9399 - mean_squared_error: 85.9399 - val_loss: 25.1000 - val_mean_squared_error: 25.1000
Epoch 13/100
81/81 [==============================] - 3s 40ms/step - loss: 81.7672 - mean_squared_error: 81.7672 - val_loss: 25.0967 - val_mean_squared_error: 25.0967
Epoch 14/100
81/81 [==============================] - 3s 40ms/step - loss: 82.4506 - mean_squared_error: 82.4506 - val_loss: 44.3703 - val_mean_squared_error: 44.3703
Epoch 15/100
81/81 [==============================] - 3s 40ms/step - loss: 66.9607 - mean_squared_error: 66.9607 - val_loss: 25.9203 - val_mean_squared_error: 25.9203
Epoch 16/100
81/81 [==============================] - 3s 40ms/step - loss: 71.2998 - mean_squared_error: 71.2998 - val_loss: 25.1644 - val_mean_squared_error: 25.1644
Epoch 17/100
81/81 [==============================] - 3s 40ms/step - loss: 73.1835 - mean_squared_error: 73.1835 - val_loss: 51.8107 - val_mean_squared_error: 51.8107
Epoch 18/100
81/81 [==============================] - 3s 40ms/step - loss: 67.7575 - mean_squared_error: 67.7575 - val_loss: 46.3968 - val_mean_squared_error: 46.3968
Epoch 19/100
81/81 [==============================] - 3s 40ms/step - loss: 83.6338 - mean_squared_error: 83.6338 - val_loss: 43.9463 - val_mean_squared_error: 43.9463
Epoch 20/100
81/81 [==============================] - 3s 40ms/step - loss: 119.4474 - mean_squared_error: 119.4474 - val_loss: 126.3989 - val_mean_squared_error: 126.3989
Epoch 21/100
81/81 [==============================] - 3s 40ms/step - loss: 70.1048 - mean_squared_error: 70.1048 - val_loss: 28.8203 - val_mean_squared_error: 28.8203
Epoch 22/100
81/81 [==============================] - 3s 40ms/step - loss: 64.7738 - mean_squared_error: 64.7738 - val_loss: 95.8141 - val_mean_squared_error: 95.8141
Epoch 23/100
81/81 [==============================] - 3s 41ms/step - loss: 68.6635 - mean_squared_error: 68.6635 - val_loss: 37.7457 - val_mean_squared_error: 37.7457
Epoch 24/100
81/81 [==============================] - 3s 42ms/step - loss: 60.5811 - mean_squared_error: 60.5811 - val_loss: 28.6782 - val_mean_squared_error: 28.6782
Epoch 25/100
81/81 [==============================] - 4s 43ms/step - loss: 60.8526 - mean_squared_error: 60.8526 - val_loss: 34.1834 - val_mean_squared_error: 34.1834
Epoch 26/100
81/81 [==============================] - 3s 41ms/step - loss: 58.8444 - mean_squared_error: 58.8444 - val_loss: 25.5026 - val_mean_squared_error: 25.5026
Epoch 27/100
81/81 [==============================] - 3s 40ms/step - loss: 53.5484 - mean_squared_error: 53.5484 - val_loss: 20.1564 - val_mean_squared_error: 20.1564
Epoch 28/100
81/81 [==============================] - 3s 40ms/step - loss: 56.3856 - mean_squared_error: 56.3856 - val_loss: 57.2569 - val_mean_squared_error: 57.2569
Epoch 29/100
81/81 [==============================] - 3s 41ms/step - loss: 52.0987 - mean_squared_error: 52.0987 - val_loss: 29.2898 - val_mean_squared_error: 29.2898
Epoch 30/100
81/81 [==============================] - 3s 40ms/step - loss: 54.8269 - mean_squared_error: 54.8269 - val_loss: 25.0660 - val_mean_squared_error: 25.0660
Epoch 31/100
81/81 [==============================] - 3s 40ms/step - loss: 47.5326 - mean_squared_error: 47.5326 - val_loss: 25.7464 - val_mean_squared_error: 25.7464
Epoch 32/100
81/81 [==============================] - 3s 40ms/step - loss: 49.9826 - mean_squared_error: 49.9826 - val_loss: 38.5616 - val_mean_squared_error: 38.5616
Epoch 33/100
81/81 [==============================] - 3s 40ms/step - loss: 51.8281 - mean_squared_error: 51.8281 - val_loss: 51.8559 - val_mean_squared_error: 51.8559
Epoch 34/100
81/81 [==============================] - 3s 41ms/step - loss: 44.0718 - mean_squared_error: 44.0718 - val_loss: 69.3781 - val_mean_squared_error: 69.3781
Epoch 35/100
81/81 [==============================] - 3s 42ms/step - loss: 41.5928 - mean_squared_error: 41.5928 - val_loss: 51.8529 - val_mean_squared_error: 51.8529
Epoch 36/100
81/81 [==============================] - 3s 41ms/step - loss: 50.8295 - mean_squared_error: 50.8295 - val_loss: 26.3513 - val_mean_squared_error: 26.3513
Epoch 37/100
81/81 [==============================] - 3s 41ms/step - loss: 48.1331 - mean_squared_error: 48.1331 - val_loss: 51.0451 - val_mean_squared_error: 51.0451
Epoch 38/100
81/81 [==============================] - 3s 42ms/step - loss: 51.8168 - mean_squared_error: 51.8168 - val_loss: 78.9762 - val_mean_squared_error: 78.9762
Epoch 39/100
81/81 [==============================] - 3s 42ms/step - loss: 43.3375 - mean_squared_error: 43.3375 - val_loss: 30.0821 - val_mean_squared_error: 30.0821
Epoch 40/100
81/81 [==============================] - 3s 41ms/step - loss: 45.3207 - mean_squared_error: 45.3207 - val_loss: 126.6301 - val_mean_squared_error: 126.6301
Epoch 41/100
81/81 [==============================] - 3s 41ms/step - loss: 42.8571 - mean_squared_error: 42.8571 - val_loss: 23.0032 - val_mean_squared_error: 23.0032
Epoch 42/100
81/81 [==============================] - 3s 42ms/step - loss: 47.0686 - mean_squared_error: 47.0686 - val_loss: 23.2250 - val_mean_squared_error: 23.2250
Epoch 43/100
81/81 [==============================] - 3s 41ms/step - loss: 45.5567 - mean_squared_error: 45.5567 - val_loss: 94.6416 - val_mean_squared_error: 94.6416
Epoch 44/100
81/81 [==============================] - 3s 40ms/step - loss: 42.3307 - mean_squared_error: 42.3307 - val_loss: 118.0972 - val_mean_squared_error: 118.0972
Epoch 45/100
81/81 [==============================] - 3s 41ms/step - loss: 42.6321 - mean_squared_error: 42.6321 - val_loss: 62.2360 - val_mean_squared_error: 62.2360
Epoch 46/100
81/81 [==============================] - 3s 40ms/step - loss: 39.0334 - mean_squared_error: 39.0334 - val_loss: 124.5826 - val_mean_squared_error: 124.5826
Epoch 47/100
81/81 [==============================] - 3s 40ms/step - loss: 38.0613 - mean_squared_error: 38.0613 - val_loss: 121.8002 - val_mean_squared_error: 121.8002
Epoch 48/100
81/81 [==============================] - 3s 40ms/step - loss: 38.9757 - mean_squared_error: 38.9757 - val_loss: 54.3880 - val_mean_squared_error: 54.3880
Epoch 49/100
81/81 [==============================] - 3s 40ms/step - loss: 43.1675 - mean_squared_error: 43.1675 - val_loss: 32.4478 - val_mean_squared_error: 32.4478
Epoch 50/100
81/81 [==============================] - 3s 41ms/step - loss: 41.5491 - mean_squared_error: 41.5491 - val_loss: 36.0595 - val_mean_squared_error: 36.0595
Epoch 51/100
81/81 [==============================] - 3s 41ms/step - loss: 43.8271 - mean_squared_error: 43.8271 - val_loss: 21.4376 - val_mean_squared_error: 21.4376
Epoch 52/100
81/81 [==============================] - 3s 41ms/step - loss: 42.3230 - mean_squared_error: 42.3230 - val_loss: 45.6934 - val_mean_squared_error: 45.6934
Epoch 53/100
81/81 [==============================] - 3s 42ms/step - loss: 37.9259 - mean_squared_error: 37.9259 - val_loss: 33.1510 - val_mean_squared_error: 33.1510
Epoch 54/100
81/81 [==============================] - 3s 41ms/step - loss: 39.8762 - mean_squared_error: 39.8762 - val_loss: 124.4165 - val_mean_squared_error: 124.4165
Epoch 55/100
81/81 [==============================] - 3s 42ms/step - loss: 36.5833 - mean_squared_error: 36.5833 - val_loss: 74.2720 - val_mean_squared_error: 74.2720
Epoch 56/100
81/81 [==============================] - 4s 44ms/step - loss: 39.4541 - mean_squared_error: 39.4541 - val_loss: 37.1163 - val_mean_squared_error: 37.1163
Epoch 57/100
81/81 [==============================] - 3s 42ms/step - loss: 38.4480 - mean_squared_error: 38.4480 - val_loss: 20.9749 - val_mean_squared_error: 20.9749
Epoch 58/100
81/81 [==============================] - 3s 42ms/step - loss: 46.1069 - mean_squared_error: 46.1069 - val_loss: 161.2850 - val_mean_squared_error: 161.2849
Epoch 59/100
81/81 [==============================] - 3s 41ms/step - loss: 43.5871 - mean_squared_error: 43.5871 - val_loss: 66.5388 - val_mean_squared_error: 66.5388
Epoch 60/100
81/81 [==============================] - 3s 41ms/step - loss: 42.5840 - mean_squared_error: 42.5840 - val_loss: 156.7058 - val_mean_squared_error: 156.7058
Epoch 61/100
81/81 [==============================] - 3s 42ms/step - loss: 36.2576 - mean_squared_error: 36.2576 - val_loss: 138.9029 - val_mean_squared_error: 138.9029
Epoch 62/100
81/81 [==============================] - 3s 41ms/step - loss: 37.9584 - mean_squared_error: 37.9584 - val_loss: 93.8942 - val_mean_squared_error: 93.8942
Epoch 63/100
81/81 [==============================] - 3s 41ms/step - loss: 42.2680 - mean_squared_error: 42.2680 - val_loss: 69.3455 - val_mean_squared_error: 69.3455
Epoch 64/100
81/81 [==============================] - 3s 42ms/step - loss: 34.3788 - mean_squared_error: 34.3788 - val_loss: 25.6917 - val_mean_squared_error: 25.6917
Epoch 65/100
81/81 [==============================] - 3s 42ms/step - loss: 36.9716 - mean_squared_error: 36.9716 - val_loss: 95.5516 - val_mean_squared_error: 95.5516
Epoch 66/100
81/81 [==============================] - 3s 42ms/step - loss: 37.3081 - mean_squared_error: 37.3081 - val_loss: 74.6094 - val_mean_squared_error: 74.6094
Epoch 67/100
81/81 [==============================] - 3s 42ms/step - loss: 39.0882 - mean_squared_error: 39.0882 - val_loss: 23.3814 - val_mean_squared_error: 23.3814
Epoch 68/100
81/81 [==============================] - 3s 42ms/step - loss: 40.4898 - mean_squared_error: 40.4898 - val_loss: 141.8593 - val_mean_squared_error: 141.8593
Epoch 69/100
81/81 [==============================] - 3s 41ms/step - loss: 34.6456 - mean_squared_error: 34.6456 - val_loss: 45.9083 - val_mean_squared_error: 45.9083
Epoch 70/100
81/81 [==============================] - 3s 41ms/step - loss: 35.1703 - mean_squared_error: 35.1703 - val_loss: 187.8447 - val_mean_squared_error: 187.8447
Epoch 71/100
81/81 [==============================] - 3s 43ms/step - loss: 32.8633 - mean_squared_error: 32.8633 - val_loss: 45.8127 - val_mean_squared_error: 45.8127
Epoch 72/100
81/81 [==============================] - 3s 43ms/step - loss: 39.1667 - mean_squared_error: 39.1667 - val_loss: 108.1412 - val_mean_squared_error: 108.1412
Epoch 73/100
81/81 [==============================] - 3s 42ms/step - loss: 175.9664 - mean_squared_error: 175.9664 - val_loss: 2688.0950 - val_mean_squared_error: 2688.0950
Epoch 74/100
81/81 [==============================] - 3s 43ms/step - loss: 55.3466 - mean_squared_error: 55.3466 - val_loss: 175.4979 - val_mean_squared_error: 175.4979
Epoch 75/100
81/81 [==============================] - 3s 43ms/step - loss: 61.4937 - mean_squared_error: 61.4937 - val_loss: 32.3055 - val_mean_squared_error: 32.3055
Epoch 76/100
81/81 [==============================] - 3s 42ms/step - loss: 46.5618 - mean_squared_error: 46.5618 - val_loss: 124.7615 - val_mean_squared_error: 124.7615
Epoch 77/100
81/81 [==============================] - 3s 43ms/step - loss: 45.6146 - mean_squared_error: 45.6146 - val_loss: 207.3768 - val_mean_squared_error: 207.3768
Epoch 78/100
81/81 [==============================] - 3s 42ms/step - loss: 42.4978 - mean_squared_error: 42.4978 - val_loss: 57.7087 - val_mean_squared_error: 57.7087
Epoch 79/100
81/81 [==============================] - 3s 41ms/step - loss: 39.5786 - mean_squared_error: 39.5786 - val_loss: 47.5629 - val_mean_squared_error: 47.5629
Epoch 80/100
81/81 [==============================] - 3s 42ms/step - loss: 43.1608 - mean_squared_error: 43.1608 - val_loss: 128.4153 - val_mean_squared_error: 128.4153
Epoch 81/100
81/81 [==============================] - 3s 42ms/step - loss: 45.6360 - mean_squared_error: 45.6360 - val_loss: 26.3359 - val_mean_squared_error: 26.3359
Epoch 82/100
81/81 [==============================] - 3s 42ms/step - loss: 34.5775 - mean_squared_error: 34.5775 - val_loss: 37.1616 - val_mean_squared_error: 37.1616
Epoch 83/100
81/81 [==============================] - 3s 42ms/step - loss: 39.1843 - mean_squared_error: 39.1843 - val_loss: 100.8166 - val_mean_squared_error: 100.8166
Epoch 84/100
81/81 [==============================] - 3s 42ms/step - loss: 40.6150 - mean_squared_error: 40.6150 - val_loss: 116.2442 - val_mean_squared_error: 116.2442
Epoch 85/100
81/81 [==============================] - 3s 43ms/step - loss: 32.1132 - mean_squared_error: 32.1132 - val_loss: 130.3223 - val_mean_squared_error: 130.3223
Epoch 86/100
81/81 [==============================] - 4s 44ms/step - loss: 37.9627 - mean_squared_error: 37.9627 - val_loss: 33.6156 - val_mean_squared_error: 33.6156
Epoch 87/100
81/81 [==============================] - 3s 42ms/step - loss: 36.4648 - mean_squared_error: 36.4648 - val_loss: 69.2918 - val_mean_squared_error: 69.2918
Epoch 88/100
81/81 [==============================] - 3s 41ms/step - loss: 36.2454 - mean_squared_error: 36.2454 - val_loss: 42.8555 - val_mean_squared_error: 42.8555
Epoch 89/100
81/81 [==============================] - 4s 44ms/step - loss: 37.7223 - mean_squared_error: 37.7223 - val_loss: 238.7253 - val_mean_squared_error: 238.7253
Epoch 90/100
81/81 [==============================] - 4s 43ms/step - loss: 33.2982 - mean_squared_error: 33.2982 - val_loss: 85.5239 - val_mean_squared_error: 85.5239
Epoch 91/100
81/81 [==============================] - 3s 42ms/step - loss: 32.3676 - mean_squared_error: 32.3676 - val_loss: 31.6639 - val_mean_squared_error: 31.6639
Epoch 92/100
81/81 [==============================] - 3s 42ms/step - loss: 29.3923 - mean_squared_error: 29.3923 - val_loss: 176.0966 - val_mean_squared_error: 176.0966
Epoch 93/100
81/81 [==============================] - 3s 41ms/step - loss: 35.7773 - mean_squared_error: 35.7773 - val_loss: 178.3221 - val_mean_squared_error: 178.3221
Epoch 94/100
81/81 [==============================] - 3s 41ms/step - loss: 34.2187 - mean_squared_error: 34.2187 - val_loss: 82.0698 - val_mean_squared_error: 82.0698
Epoch 95/100
81/81 [==============================] - 3s 43ms/step - loss: 31.9256 - mean_squared_error: 31.9256 - val_loss: 91.5862 - val_mean_squared_error: 91.5862
Epoch 96/100
81/81 [==============================] - 3s 43ms/step - loss: 29.1709 - mean_squared_error: 29.1709 - val_loss: 129.5807 - val_mean_squared_error: 129.5807
Epoch 97/100
81/81 [==============================] - 3s 42ms/step - loss: 30.8514 - mean_squared_error: 30.8514 - val_loss: 66.6177 - val_mean_squared_error: 66.6177
Epoch 98/100
81/81 [==============================] - 3s 42ms/step - loss: 35.3912 - mean_squared_error: 35.3912 - val_loss: 39.7499 - val_mean_squared_error: 39.7499
Epoch 99/100
81/81 [==============================] - 3s 41ms/step - loss: 32.4280 - mean_squared_error: 32.4280 - val_loss: 54.4770 - val_mean_squared_error: 54.4770
Epoch 100/100
81/81 [==============================] - 3s 41ms/step - loss: 32.3317 - mean_squared_error: 32.3317 - val_loss: 134.0100 - val_mean_squared_error: 134.0100


y_pred__test_ga_lstm_f = model_lstm.predict(test_data_ga_ann_f)
y_pred__train_ga_lstm_f = model_lstm.predict(train_data_ga_ann_f)

2022-12-10 17:51:44.796456: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.
2022-12-10 17:51:44.916820: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.
2022-12-10 17:51:44.984218: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.
2022-12-10 17:51:45.076707: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:112] Plugin optimizer for device_type GPU is enabled.


plt.figure(figsize = (12,6), dpi = 400)
plt.plot(y_train_ann, label = 'Actual')
#plt.plot(y_pred__train_ga_nn, label = 'GA-ANN-F')
#plt.plot(y_pred__train_ann, label = 'ANN')
plt.plot(y_pred__train_lstm, label = 'LSTM')
plt.plot(y_pred__train_ga_lstm_f, label = 'GA-LSTM-F')
plt.legend()

<matplotlib.legend.Legend at 0x2c441c6d0>


plt.figure(figsize = (12,6), dpi = 400)
plt.plot(y_test, label = 'Actual')
#plt.plot(y_pred__train_ga_nn, label = 'GA-ANN-F')
#plt.plot(y_pred__train_ann, label = 'ANN')
plt.plot(y_pred__test_lstm, label = 'LSTM')
plt.plot(y_pred__test_ga_lstm_f, label = 'GA-LSTM-F')
plt.legend()

<matplotlib.legend.Legend at 0x28f6b2fd0>


arr_lstm_ga_f = list(y_pred__train_ga_lstm_f) + list(y_pred__test_ga_lstm_f)
arr_lstm = list(y_pred__train_lstm) + list(y_pred__test_lstm)


arr_lstm_ga_f1 = []
for x in arr_lstm_ga_f:
    arr_lstm_ga_f1.append(float(x))
    
arr_lstm1 = []
for x in arr_lstm:
    arr_lstm1.append(float(x))


temp = pd.read_csv('temp1.csv', index_col='Unnamed: 0')


temp['LSTM'] = arr_lstm1
temp['GA_LSTM_F'] = arr_lstm_ga_f1


temp


#temp.to_csv('temp1.csv')


temp['DATE'] = pd.to_datetime(temp['DATE'])


plt.figure(figsize = (12,6), dpi = 600)

plt.plot( temp['DATE'],temp['Actual'], label = 'Actual')
plt.plot( temp['DATE'],temp['ANN'], label = 'ANN')
plt.plot( temp['DATE'],temp['GA_ANN_F'], label = 'GA-ANN-F')
plt.plot( temp['DATE'],temp['LSTM'], label = 'LSTM')
plt.plot( temp['DATE'],temp['GA_LSTM_F'], label = 'GA-ANN-F')

plt.vlines(x = datetime(2018,8,18), ymin = 0, ymax= 240, color = 'red', linewidth = 2)
#plt.plot(predictions_ga_ann_test, label = 'ANN_GA',  marker = '*', alpha  =  0.5)
#plt.plot(predictions_ga_ann_test, label = 'LSTM_GA', alpha  =  0.5)
plt.legend()
plt.grid(linestyle = '-.', alpha = 0.5)
plt.ylim(0,160)
#plt.savefig('PlotFinal.jpg', dpi = 800)
plt.show()

pl

	DATE	Actual	ANN	GA_ANN_F
14	2007-10-08	86.19	37.727489	64.785660
15	2007-10-09	87.58	39.110336	64.490265
16	2007-10-10	87.19	38.735291	64.919067
17	2007-10-11	89.48	40.521767	66.621346
18	2007-10-12	88.58	42.567089	64.885735
...	...	...	...	...
2914	2020-02-13	51.36	25.968416	45.890278
2915	2020-02-14	49.78	26.022779	46.879910
2916	2020-02-18	48.67	26.122488	45.461826
2917	2020-02-19	47.17	27.231556	46.597939
2918	2020-02-20	44.83	28.258421	46.854820

	DATE	Actual	ANN	GA_ANN_F	LSTM	GA_LSTM_F
14	2007-10-08	86.19	37.727489	64.785660	70.849083	89.754662
15	2007-10-09	87.58	39.110336	64.490265	71.228577	89.821960
16	2007-10-10	87.19	38.735291	64.919067	72.074455	89.924789
17	2007-10-11	89.48	40.521767	66.621346	72.689896	90.033119
18	2007-10-12	88.58	42.567089	64.885735	73.109650	90.113106
...	...	...	...	...	...	...
2914	2020-02-13	51.36	25.968416	45.890278	50.374458	71.115837
2915	2020-02-14	49.78	26.022779	46.879910	50.724934	71.307602
2916	2020-02-18	48.67	26.122488	45.461826	51.769032	72.148445
2917	2020-02-19	47.17	27.231556	46.597939	52.022560	72.775902
2918	2020-02-20	44.83	28.258421	46.854820	52.072510	72.981224